Humanoid robots exhibit significant potential in executing diverse human-level skills. However, current research predominantly relies on data-driven approaches that necessitate extensive training datasets to achieve robust multimodal decision-making capabilities and generalizable visuomotor control. These methods raise concerns due to the neglect of geometric reasoning in unseen scenarios and the inefficient modeling of robot-target relationships within the training data, resulting in a significant waste of training resources. To address these limitations, we present the Recurrent Geometric-prior Multimodal Policy (RGMP), an end-to-end framework that unifies geometric-semantic skill reasoning with data-efficient visuomotor control. For perception capabilities, we propose the Geometric-prior Skill Selector, which infuses geometric inductive biases into a vision language model, producing adaptive skill sequences for unseen scenes with minimal spatial common sense tuning. To achieve data-efficient robotic motion synthesis, we introduce the Adaptive Recursive Gaussian Network, which parameterizes robot-object interactions as a compact hierarchy of Gaussian processes that recursively encode multi-scale spatial relationships, yielding dexterous, data-efficient motion synthesis even from sparse demonstrations. Evaluated on both our humanoid robot and desktop robot, the RGMP framework achieves 87% task success in generalization tests and exhibits 5× greater data efficiency than the state-of-the-art model. This performance underscores its superior cross-domain generalization, paving the way for more versatile and data-efficient robotic systems.

The massive scale of data and computation required for training Multimodal Large Language Models (MLLMs) has fueled the rise of Fine-Tuning as a Service (FTaaS), enabling users to rapidly customize models for diverse real-world tasks. While FTaaS democratizes access to advanced multimodal intelligence, it also introduces serious security concerns, particularly backdoor attacks. In this work, we systematically analyze backdoor vulnerabilities in MLLMs under the FTaaS paradigm, revealing two key phenomena: (1) markedly reduced sensitivity to textual variations when a visual trigger is present, and (2) abnormally stable model confidence even under strong semantic perturbations. Building on these insights, we propose Trap on Text (ToT), a novel inference-time backdoor detection framework. ToT applies controlled semantic perturbations to textual prompts and jointly analyzes the semantic consistency and confidence drift of the model’s responses, enabling robust detection of backdoor activations without requiring model parameters, architectures or clean reference data. Extensive experiments across architectures and datasets show that ToT achieves strong attack mitigation and preserves clean accuracy, offering a practical solution for safeguarding FTaaS workflows.

Text-Attributed Graphs (TAGs) are graphs where both nodes and edges are associated with text attributes. To leverage their semantic richness, recent efforts have integrated large language models (LLMs) with graph neural networks, leading to the development of GraphLLMs. However, many real-world datasets remain inaccessible, and processing text-attributed graphs while ensuring privacy and efficiency remains a challenge. To address this, we place TAGs within a federated environment, referred to as TAG-FGL. Despite its potential, TAG-FGL remains largely underexplored in the face of adversarial threats. In this work, we introduce GTAE, a novel attack framework that cascades influence-guided topological perturbations and embedding-level text refinements to generate transferable, modality-agnostic adversarial inputs. To defend against these threats, we propose STRUM, a defense strategy that combines local adversarial training with robustness-aware aggregation, enhancing resilience at both the node and system levels. Extensive experiments on five real-world datasets with diverse model backbones demonstrate that GTAE significantly degrades model performance, while STRUM consistently improves robustness.

Federated learning enables multiple medical institutions to train a global model without sharing data, yet feature heterogeneity from diverse scanners or protocols remains a major challenge. Many existing works attempt to address this issue by leveraging model representations (e.g., mean feature vectors) to correct local training; however, they often face two key limitations: 1) Incomplete Contextual Representation Learning: Current approaches primarily focus on final-layer features, overlooking critical multi-level cues and thus diluting essential context for accurate segmentation. 2) Layerwise Style Bias Accumulation: Although utilizing representations can partially align global features, these methods neglect domain-specific biases within intermediate layers, allowing style discrepancies to build up and reduce model robustness. To address these challenges, we propose FedBCS to bridge feature representation gaps via domain-invariant contextual prototypes alignment. Specifically, we introduce a frequency-domain adaptive style recalibration into prototype construction that not only decouples content-style representations but also learns optimal style parameters, enabling more robust domain-invariant prototypes. Furthermore, we design a context-aware dual-level prototype alignment method that extracts domain-invariant prototypes from different layers of both encoder and decoder and fuses them with contextual information for finer-grained representation alignment. Extensive experiments on two public datasets demonstrate that our method exhibits remarkable performance.

This work introduces Text-based Aerial-Ground Person Retrieval (TAG-PR), which aims to retrieve person images from heterogeneous aerial and ground views with textual descriptions. Unlike traditional Text-based Person Retrieval (T-PR), which focuses solely on ground-view images, TAG-PR introduces greater practical significance and presents unique challenges due to the large viewpoint discrepancy across images. To support this task, we contribute: (1) TAG-PEDES dataset, constructed from public benchmarks with automatically generated textual descriptions, enhanced by a diversified text generation paradigm to ensure robustness under view heterogeneity; and (2) TAG-CLIP, a novel retrieval framework that addresses view heterogeneity through a hierarchically-routed mixture of experts module to learn view-specific and view-agnostic features and a viewpoint decoupling strategy to decouple view-specific features for better cross-modal alignment. We evaluate the effectiveness of TAG-CLIP on both the proposed TAG-PEDES and existing T-PR benchmarks.

Federated Graph Learning (FGL) has emerged as a compelling paradigm for distributed Graph Neural Networks (GNNs) training, prioritizing data privacy preservation. However, due to the limitations of data collection and storage conditions, FGL suffers from data corruption in real-world applications. While Federated Learning (FL) and FGL studies have addressed label corruption, the challenge of graph topological corruption remains unexamined. Specifically, this phenomenon significantly disrupts node connectivity patterns of graphs, leading GNNs to adopt flawed feature propagation paradigms. Existing methods with poor robustness are inevitably constrained due to the absence of targeted strategies for addressing the issues of global contaminated collaboration and local vulnerability. To tackle this challenge, we conduct the first comprehensive investigation of robust FGL against topological corruption and propose FedPRE. It comprises: (1) Feature Propagation Robustness Evaluation (FPRE), which evaluates client GNNs feature propagation robustness and adjusts their contribution during aggregation. (2) Topological Corruption-Resistant Enhancement (TCRE), which enhances robustness against corruption during local training. Extensive experiments validate the robustness and effectiveness of FedPRE against topological corruption. The code is available at https://github.com/OakleyTan/FedPRE.

Text-to-image person retrieval (TIPR) aims to identify the target person using textual descriptions, facing challenge in modality heterogeneity. Prior works have attempted to address it by developing cross-modal global or local alignment strategies. However, global methods typically overlook fine-grained cross-modal differences, whereas local methods require prior information to explore explicit part alignments. Additionally, current methods are English-centric, restricting their application in multilingual contexts. To alleviate these issues, we pioneer a multilingual TIPR task by developing a multilingual TIPR benchmark, for which we leverage large language models for initial translations and refine them by integrating domain-specific knowledge. Correspondingly, we propose Bi-IRRA: a Bidirectional Implicit Relation Reasoning and Aligning framework to learn alignment across languages and modalities. Within Bi-IRRA, a bidirectional implicit relation reasoning module enables bidirectional prediction of masked image and text, implicitly enhancing the modeling of local relations across languages and modalities, a multi-dimensional global alignment module is integrated to bridge the modality heterogeneity. The proposed method achieves new state-of-the-art results on all multilingual TIPR datasets.

Two-view correspondence pruning aims to identify reliable correspondences for camera pose estimation, serving as a fundamental step in many 3D vision tasks. Existing methods rely on geometric consistency to seek true correspondences (inliers) from numerous false correspondences (outliers). In this learning paradigm, outliers severely affect the representation learning of inliers, resulting in models that are neither robust nor generalizable. To address this issue, we propose a geometry-consistent pre-training paradigm that sculpts scalable and generalizable representations free from outlier interference. The paradigm features two appealing properties. 1) Implementation of geometry-consistent pre-training. We introduce masked inlier reconstruction as a pretext task and develop a simple yet effective pre-training framework based on a masked autoencoder. Specifically, due to the irregular and unordered nature of correspondences, which lack explicit positional information, we adopt a dual-branch structure that separately reconstructs the keypoints of two images. This enables indirect reconstruction of 4D correspondences, where keypoints from the paired image provide positional prompts. 2) Unified correspondence encoder. We propose a simple dual-stream encoder with built-in consensus interaction, providing a unified, extensible architecture that enhances representation learning. Extensive experiments demonstrate that our method, GeneralPruner, consistently outperforms state-of-the-art approaches in terms of robustness and generalization across various downstream tasks. Specifically, our method achieves 10.76%, 11.84%, and 8.65% performance gains in camera pose estimation, visual localization, and 3D registration, respectively. To the best of our knowledge, we are the first work to introduce a pre-training framework tailored for correspondence pruning, offering a more universal and scalable solution.

Pedestrian behavior exhibits inherent multi-modality, necessitating predictions that balance accuracy and diversity to adapt effectively to various complex scenarios. However, conventional noise addition in diffusion models is often aimless and unguided, leading to redundant noise reduction steps and the generation of uncontrollable samples. To address these issues, we propose a Prior Condition-Guided Diffusion Model (CGD-TraP) for multi-modal pedestrian trajectory prediction. Instead of directly adding Gaussian noise to trajectories at each timestep during the forward process, our approach leverages internal intention and external interaction to guide noise estimation. Specifically, we design two specialized modules to extract and aggregate intention and interaction features. These features are then adaptively fused through a spatial-temporal fusion based on selective state space, which estimates a controllable noisy trajectory distribution. By optimizing the noise addition process in a more controlled and efficient manner, our method ensures that the denoising process is effectively guided, resulting in predictions that are both accurate and diverse. Extensive experiments on the ETH-UCY, SDD, and NBA datasets demonstrate that CGD-TraP surpasses state-of-the-art diffusion-based and other generative methods, achieving superior efficiency, accuracy, and diversity.

Real-world person re-identification (Re-ID) systems are susceptible to malicious attacks, leading to the leakage of pedestrian images and the Re-ID model, posing severe threats to the privacy of both system owners and pedestrians. Existing privacy-preserving person re-identification (PPPR) methods fail to simultaneously resist data leakage, model leakage, and data & model leakage while compromising the normal functionality of Re-ID systems. In this paper, we begin with an in-depth analysis of prior methodologies and identify the gap between existing works and the ideal PPPR paradigm. Inspired by the concept of "Let the invisible perturbation become the system trigger", we propose SHIELD, a pioneering and comprehensive two-stage privacy-preserving framework. To resist data leakage, we propose a self-supervised method for Protected Dataset Generation in the first stage, which obviates the dependence on identity labels and ensures image quality. To resist model leakage without compromising the normal retrieval accuracy, we propose Original Feature Deconstruction and Protected Feature Alignment to train the system model with paired protected and original images. Extensive experiments substantiate that SHIELD significantly outperforms existing PPPR methods, offering robust and holistic protection for Re-ID systems while maintaining decent retrieval accuracy for authorized users. The code will be released soon.

Split Learning (SL) is a distributed learning framework that has gained popularity for its privacy-preserving nature and low computational demands. However, recent studies have the potential that a server adversary to carry out inference attacks, compromising the privacy of victim clients. Nevertheless, upon re-evaluating prior studies, we found that existing methods rely on overly strong assumptions to enhance their performance, resulting in a significant decline in effectiveness under more realistic scenarios. In this work, we provide new insights into the inherent vulnerabilities of SL. Specifically, we discover that both the smashed data and the server model contain the client's representation preference, which the server adversary can exploit to build a substitute client that approximates the target client's unique feature extraction behavior. With a well-trained substitute client, the server can perfectly steal the target client's functionality, training data, and labels. Building on this observation, we introduce Split Leakage (SLeak), a new threat that targets multiple privacy stealing objectives against SL. Notably, SLeak does not depend on strong privacy priors and only requires partial same-domain auxiliary public data to conduct the attacks. Experimental results on diverse datasets and target models show that SLeak surpasses the state-of-the-art method across multiple metrics. Moreover, ablation studies further confirm its robustness and applicability under various scenarios and assumptions.

Holistic Visual Understanding (HVU), encompassing tasks like intention recognition, emotion analysis, scene understanding, and content moderation, necessitates integrating low-level visual perception ('sight') with high-level semantic reasoning ('semantics'). While large Vision-Language Models (VLMs) like CLIP offer powerful representations, their inherent 'sight' bias limits their direct application to these semantically rich tasks. Our prior work, IntCLIP, addressed Multi-label Intention Understanding (MIU) using a dual-branch architecture but faced challenges with label generation instability (Hierarchical Class Integration - HCI) and limited feature interaction (unidirectional Sight-assisted Aggregation). This paper introduces an enhanced framework that significantly extends IntCLIP to tackle the broader HVU challenge. We propose Semantic Label Refinement (SLR), an iterative, metric-guided process leveraging Large Language Models (LLMs) and quantitative evaluation within the CLIP embedding space to generate stable, optimized semantic labels. We also introduce a novel bidirectional attention mechanism (Symmetric Aggregation) that enables balanced, mutual refinement between sight and semantic feature maps. By evaluating on a comprehensive benchmark spanning MIU, Image Emotion Recognition, Indoor Scene Recognition, and Visual Content Moderation, we demonstrate that our framework not only advances the state-of-the-art in MIU but also achieves superior performance across diverse HVU tasks. This framework provides a unified and robust solution for synergizing sight and semantics, pushing towards more human-like visual intelligence. Code is available at https://github.com/yan9qu/PAMI25-HVU.

Federated learning (FL) enables multiple clients to learn collaboratively in a distributed way, allowing for privacy protection. However, the real-world nonindependent and identically distributed (non-IID) data will lead to client drift, which degrades the performance of FL. Interestingly, we find that the logit difference between the local and global models increases as the model is continuously updated, which is the primary factor behind performance degradation. This is mainly due to catastrophic forgetting caused by non-IID data between clients. To alleviate this problem, we propose a new algorithm, named FedCSD, a class-prototype similarity distillation in a federated framework to align the logits of local and global models. FedCSD does not simply transfer global knowledge to local clients, as an insufficiently trained global model cannot provide reliable knowledge, i.e., class similarity information, and its wrong soft labels will mislead the optimization of local models. Concretely, FedCSD leverages the similarity between local logits and the global prototype to refine the global logits, thereby enhancing its class similarity information. Furthermore, FedCSD adopts an adaptive mask to filter out the terrible soft labels of the global models, thereby preventing them from misleading local optimization. Extensive experiments demonstrate the superiority of our method over the state-of-the-art FL approaches in various non-IID settings. Code is publicly available at https://github.com/IAMJackYan/FedCSD.

Single-positive multi-label learning (SPMLL) aims to train a multi-label classifier from data with single-positive label, to predict all applicable labels during testing. However, existing SPMLL methods are tailored for centralized datasets, which fail to be directly deployed to distributed setting like federated learning. In this paper, we start the first attempt to study federated single-positive multi-label learning (FedSPMLL), aiming to collaboratively train a SPMLL model from distributed data. To achieve this, we need to address challenges caused by label incompleteness: limited generalization ability of local model and overweighting contribution of client with local dataset suffering from severe label incompleteness. To this end, we propose a novelFedLOGmethod, guidingFedSPMLL with predicateLOGic-modeled label correlation. Enabling the informative knowledge extraction from limited data, we propose to model label correlation within local dataset using predicate logic. To alleviate false negative label issue, we propose to transfer confident label correlation knowledge to local model by self-distillation. To downweight the contribution of unreliable client owning dataset with severe label incompleteness, we propose a new measurement of label incompleteness to adjust client contribution for a fair aggregation. We establish a comprehensive FedSPMLL benchmark. And extensive experiments demonstrate the superiority of our FedLOG method.

Weakly supervised text-based person re-identification aims to retrieve specific pedestrians based on textual descriptions without identity labels available during training. This task remains challenging due to the inherent cross-modal heterogeneity and lack of identity annotations. There is a common issue of modality gap in vision language models, which in turn affects the performance of downstream tasks such as cross-modal retrieval and multimodal clustering. Specifically, in our research and experiments, we found that there is a problem of inter-modal misalignment between image and text modalities. However, existing methods rely on mutual enhancement strategies between image and text clustering, leading to the accumulation of clustering noise and affecting the final retrieval performance. To address this issue, we propose a Consensus Labelling: Prompt-guided Clustering refinement (CLPC) framework for weakly supervised text-based person re-identification. Specifically, we introduce a textual inversion network to learn a pseudo token that captures visual context, which is then integrated into natural language sentences as personalized textual prompt. To further improve clustering quality, we introduce a Nearest Neighbor-Guided Pseudo Label Mining (NGPM) method, which uses the clusters derived from personalized textual prompts to refine the clustering of image features. Additionally, we design a Dynamic Margin Triplet (DMT) loss, where the margin is adaptively adjusted using a sigmoid-based function to enhance the model’s ability to distinguish hard negative samples. We have also introduce a Normalized Distribution Matching (NDM) loss to minimize the KL divergence between the image-text matching scores and the normalized soft matching scores. The extensive experimental results on three public datasets have demonstrated the superiority of our method. Our code is available at https://github.com/LeviWeiZhi/CLPC.

Unsupervised visible-infrared person reidentification (US-VI-ReID) seeks to learn a cross-modality retrieval model without relying on manual annotations, thereby reducing the high cost associated with labeling. Recent large-scale vision-language pre-training models, such as CLIP, have shown significant potential in enhancing pure-vision-based person re-identification. However, existing CLIP-based US-VI-ReID methods focus on independently learning semantic information within the visible and infrared modalities. These methods overlook the mismatch between the pre-training data of CLIP and the downstream cross-modality data, resulting in substantial cross-modal semantic differences. Such inconsistent semantic information, which exhibits modality discrepancies, cannot ensure the accuracy of cross-modality associations and thus hampers the performance of cross-modality learning. To address these challenges and further explore the generalizable semantic representation across modalities in CLIP, we propose a novel framework named Mining Cross-Modality Implicit Semantic Association (MCSA), which focuses on learning a modality-invariant implicit semantic space to enhance cross-modality associations and feature learning. The proposed method comprises two key modules: Modality-invariant Prompt Learning and GCNs-Driven Collaboration Alignment. Specifically, to enable CLIP to learn modality-invariant semantics, we integrate a random color augmentation branch into the visible stream for joint contrastive learning for mining generalizable semantic representations. This ensures the color generalization of the constructed implicit semantic prompts. Moreover, within the cross-modal invariant implicit semantic space, we utilize Graph Convolutional Networks (GCNs) to uncover more reliable cross-modal associations. By integrating information from images and semantic graphs, we jointly refine cross-modal correspondences, enabling the model to perform precise cross-modal feature learning. Extensive experiments conducted on the SYSU-MM01 and RegDB datasets demonstrate the effectiveness of the proposed MCSA. The source code will be released.

Cross-view object-level image geo-localization (CVOIGL) aims to locate ground/drone-view query objects in satellite imagery. This task confronts two obstacles, namely view differences from imaging platform viewpoint changes and detection ambiguities from similar objects in large-scale satellite views. Existing methods typically employ uniform feature processing across objects while overlooking query-reference cross-view differences, leading to compromised localization precision when handling structurally analogous objects with scale variations. In this paper, we propose HiSymGeo, a Hierarchical Context Symbiosis framework with dual cooperative learning, achieving cross-view representation alignment and structural ambiguity resolution. Specifically, to mitigate cross-view differences, the Diversified View Enhancer (DiVE) first incorporates context-aware query enhancement for ground/drone-view representation while constructing scale-agnostic reference enhancement in satellite views to handle scale variations. These view-specific features then undergo contrastive learning via semantic-aware matching to align query and reference representations. Furthermore, the Query-Gated Multi-Expert View Fusion (QG-MEVF) introduces dynamic expert routing via multi-scale pyramidal representations, in which a Mixture-of-Experts (MoE) inspired architecture employs query-driven gating to adaptively select scale-specific fusion expert. This differentiable routing mechanism boosts structural discrimination against analogous objects, enabling precise object localization. Extensive ablation experiments demonstrate HiSymGeo's superiority, achieving state-of-the-art effectiveness while ensuring high cross-dataset generalization. We have released our code at https://github.com/chenqi142/HiSymGeo.

One-shot Text-to-Image Person Re-Identification (One-shot TIReID) aims to construct a TIReID model using only a single labeled image-text pair per identity, along with a large pool of unlabeled person images. While supervised learning in text-to-image person re-identification has demonstrated high effectiveness, the requirement for extensive annotated data, both in terms of identities and corresponding textual descriptions, makes it impractical for large-scale camera networks. One-shot TIReID presents a promising approach to reduce the annotation burden. The primary challenge in one-shot TIReID lies in establishing consistent visual-textual correspondences across diverse viewing conditions, particularly in the absence of cross-view paired data. To address this challenge, we propose a novel progressive discrepancy learning framework, termed P-CLIP, which aims to establish a shared embedding space that is robust to view-specific biases. To achieve this goal, we dynamically construct multi-view image-text pairs based on a single labeled pair and simultaneously project the multi-view data into a unified embedding space. Specifically, we propose a Progressive Multi-View Generation method (MVG) to generate multiple noisy views from a single labeled instance for training. To mitigate cross-view ambiguities, we introduce a Cross-View Discrepancy Learning module (CDL) that leverages the discrepancies among different views to guide the learning of cross-view visual-textual correspondences. This approach effectively integrates multimodal error correction into the person re-identification domain. Furthermore, to enhance the effectiveness of visual-textual correspondence learning, we propose a Compact Cross-Modal Matching Loss (CCM), which suppresses unmatched pairs while emphasizing matched ones. Extensive experiments were conducted on three benchmark datasets, and the experimental results demonstrate the effectiveness of our proposed method. The data and codes are available at https://github.com/Itachjw/P-CLIP/tree/main.

Sketch-based Person Retrieval (SBPR) aims to identify and retrieve a target individual across non-overlapping camera views using professional sketches as queries. In practice, sketches drawn by different artists often present diverse painting styles unpredictably. The substantial style variations among sketches pose significant challenges to the stability and generalizability of SBPR models. Prior works attempt to mitigate style variations through style manipulation methods, which inevitably undermine the inherent structural relations among multiple sketch features. This leads to overfitting on existing training styles and struggles with generalizing to new, unseen sketch styles. In this paper, we introduce FreeStyle, an innovative style-inclusive framework for SBPR, built upon the foundational CLIP architecture. FreeStyle explicitly models the relations across diverse sketch styles via style consistency enhancement, enabling dynamic adaptation to both seen and unseen style variations. Specifically, Diverse Style Semantic Unification is first devised to enhance the style consistency of each identity at the semantic level by introducing objective attribute-level semantic constraints. Meanwhile, Diverse Style Feature Squeezing tackles unclear feature boundaries among identities by concentrating the intra-identity space and separating the inter-identity space, thereby strengthening style consistency at the feature representation level. Additionally, considering the feature distribution discrepancy between sketches and photos, an identity-centric cross-modal prototype alignment mechanism is introduced to facilitate identity-aware cross-modal associations and promote a compact joint embedding space. Extensive experiments validate that FreeStyle not only achieves stable performance under seen style variations but also demonstrates strong generalization to unseen sketch styles.

Person image generation has become an increasingly important problem in computer vision with broad applications in virtual try-on, digital content creation, entertainment, and human-computer interaction. Despite recent advances in diffusion-based generative models that can produce photorealistic results, existing pipelines are still constrained by rigid modality requirements. Most prior methods rely on fixed patterns such as pose-plus-appearance inputs or text-only descriptions, limiting their flexibility and controllability in practical scenarios where users may prefer different or mixed types of guidance. This lack of adaptability poses a clear barrier to deployment in real-world systems. To overcome these challenges, we propose MaPPA (Multimodal Controllable Person Image Generation with Pose and Appearance Guidance), a unified multimodal framework that leverages transformer-based latent diffusion models. The central idea is to provide composable control over multiple modalities, enabling person image generation conditioned on text prompts, reference appearance images, pose keypoints, or any combination thereof. To achieve this, we introduce a unified framework that incorporates dedicated control blocks for appearance and pose guidance, which are strategically interleaved with transformer base blocks. These control blocks are conditioned on a unified multimodal embedding that integrates heterogeneous inputs into a consistent representation, thereby supporting arbitrary modality combinations with a single unified pipeline. Another key contribution of MaPPA is a cumulative classifier-free guidance strategy that enables allowing users to independently adjust the strength of appearance and pose guidance via scalar weights from different control signals. This design allows users to adjust the relative strength of appearance versus pose guidance, providing fine-grained controllability during inference. Furthermore, to address the common problem of detail loss in latent-diffusion decoders, we propose a texture enhancement decoding (TED) strategy, which fine-tunes the VAE decoder with edge-aware reconstruction objectives. This refinement significantly alleviates texture distortion, preserving high-frequency details in clothing patterns, facial regions, and other fine structures. Extensive experiments confirm that MaPPA achieves competitive quantitative scores while providing superior perceptual quality and user preference. Unlike task-specific methods, our framework-though constrained by the capacity of the unified multimodal embedding-supports combinations of text, pose, and appearance modalities within a single pipeline, demonstrating both flexibility and practical value. The code and the corresponding model will be made publicly available to the research community.

Catastrophic forgetting, the degradation of knowledge about previously seen classes when learning new concepts from a shifting data stream, is a pitfall faced by neural network learning in open environments. Recent research on continual image classification usually relies on storing samples or prototypes to resist this forgetting. We find that during acquiring knowledge of the new classes, the features of old classes gradually disperse, which leads to confusion of features between classes and makes them difficult to discriminate. Coping with feature dispersion would be a key consideration in resisting catastrophic forgetting, which has been neglected in previous works. To this end, we try to address this issue from two perspectives. First, we propose a dispersing feature generation mechanism, which generates pseudo-features based on the pre-pooling prototypes of the old classes to simulate feature dispersion and remind the classifier to adjust the decision boundary. Second, we design a consistent alignment constraint to alleviate the severity of feature dispersion by maintaining consistency in the hidden states of different depths when aligning the current model with the previous model. Extensive experimental results on various benchmarks show the superiority of our proposed method.

Vertical Federated Learning (VFL) involves multiple clients collaborating to train a global model, with distributed features of shared samples. While it becomes a critical privacy-preserving learning paradigm, its security can be significantly compromised by backdoor attacks, where a malicious client injects a target backdoor by manipulating local data. Existing attack methods in VFL rely on the assumption that the malicious client can obtain additional knowledge about task labels, which is not applicable in VFL. In this work, we investigate a new backdoor attack paradigm in VFL, Label-Free Backdoor Attacks (LFBA), which does not require any additional task label information and is feasible in VFL settings. Specifically, while existing methods assume access to task labels or target-class samples, we demonstrate that the gradients of local embeddings reflect the semantic information of labels. It can be utilized to construct the target poison sample set. Besides, we uncover that backdoor triggers tend to be ignored and under-fitted due to the learning of original features, which hinders backdoor task optimization. To address this, we propose selectively switching poison samples to disrupt feature learning, promoting backdoor task learning while maintaining accuracy on clean data. Extensive experiments demonstrate the effectiveness of our method in various settings.

Unsupervised visible-infrared person re-identification (US-VI-ReID) seeks to match infrared and visible images of the same individual without the use of annotations. Current methods typically derive cross-modal correspondences through a single global feature matching process for generating pseudo labels and learning modality-invariant features. However, this matching approach is hindered by both intra-modality and inter-modality discrepancies, which result in imprecise measurements. As a consequence, the clustering of individuals with single global feature is often incomplete and unreliable, leading to suboptimal performance in cross-modal clustering tasks. To address these challenges and to extract cross-modality discriminative identity information, we propose a TokenMatcher, which encompasses three key components: Diverse Tokens Matching (DTM), Diverse Tokens Neighbor Learning (DTNL), and the Homogeneous Fusion (HF) Module. DTM utilizes multiple class tokens within the visual transformer framework to capture diverse embedding representations, thereby facilitating the integration of fine-grained information essential for reliable cross-modality correspondences. DTNL enhances the intra-modality and inter-modality consistency among diverse tokens by refining neighborhood sets with insights from neighboring tokens and camera information, promoting robust neighborhood learning and fostering discriminative identity information. Additionally, the HF module consolidates clusters of the same identity while effectively separating those of different identities. Extensive experiments conducted on the publicly available SYSU-MM01 and RegDB datasets demonstrate the efficacy of the proposed method.

Fatigue is a critical factor contributing to accidents in industries such as safety monitoring and engineering construction. Fatigue exhibits dynamic complexity and non-stationary characteristics, so there are many intermediate states of short-term variation between alert and fatigue. Capturing and learning the signs of these intermediate states is essential for accurate fatigue assessment. However, current fatigue detection methods primarily rely on coarse-grained labels, typically spanning minutes to hours, and commonly treat alert and fatigue as two distinctly separate distributions, overlooking the expression of intermediate states and oversimplifying the rich distribution information of fatigue types and levels, thereby limiting detection effectiveness. To address these, this paper explores a refined representation of fatigue in terms of three dimensions: time, type, and level, and proposes a Multi-Dimensional Fine-Grained Modeling for Fatigue Detection (MDFG). This introduces the SmallLoss to extract trustworthy samples, utilizes clustering to identify diverse subtypes under alert and fatigued states, and establishes base class sets in each state. Subsequently, a complete base class set containing intermediate state bases is constructed using the base class synthesis method, which achieves the expression of intermediate fatigue states from absence to presence. Finally, fatigue levels are quantified based on the matching between samples and the complete base class set. Moreover, to cope with the complex variability of fatigue states, MDFG employs meta-learning for training. MDFG achieves an Average accuracy improvement of 10.0% and 12.1% on two real datasets compared to methods that do not consider fine-grained information. Extensive experiments demonstrate that the MDFG exhibits superior robustness and stability among current fatigue detection methods.

Person re-identification (Re-ID) is crucial for intelligent surveillance systems, facilitating the identification of individuals across multiple camera views. While significant advancements have been made for daytime scenarios, ensuring reliable Re-ID performance during nighttime remains a significant challenge. Given the cost and limited accessibility of infrared cameras, we investigate a critical question: Can RGB cameras be effectively utilized for accurate Re-ID during nighttime? To address this, we introduce NightReID, a large-scale RGB Re-ID dataset collected from a real-world nighttime surveillance system. NightReID includes 1,500 identities and over 53,000 images, capturing diverse scenes with complex lighting and adverse weather conditions. This rich dataset provides a valuable benchmark for advancing nighttime Re-ID research. Moreover, we propose the Enhancement, Denoising, and Alignment (EDA) framework with two novel modules to enhance nighttime Re-ID performance. First, an unsupervised Image Enhancement and Denoising (IED) method is designed to improve the quality of nighttime images, preserving critical details while removing noise without requiring paired ground truth. Second, we introduce Data Distribution Alignment (DDA) through statistical priors, aligning the distributions between pre-training data and nighttime data to mitigate domain shift. Extensive experiments on multiple nighttime Re-ID datasets demonstrate the significance of NightReID and validate the efficacy, flexibility, and applicability of the EDA framework.

Traditional text-based person retrieval (TPR) relies on a single-shot text as query to retrieve the target person, assuming that the query completely captures the user’s search intent. However, in real-world scenarios, it can be challenging to ensure the information completeness of such single-shot text. To address this limitation, we propose chat-based person retrieval (ChatPR), a new paradigm that takes an interactive dialogue as query to perform the person retrieval, engaging the user in conversational context to progressively refine the query for accurate person retrieval. The primary challenge in ChatPR is the lack of available dialogue-image paired data. To overcome this challenge, we establish ChatPedes, the first dataset designed for ChatPR, which is constructed by leveraging large language models to automate the question generation and simulate user responses. Additionally, to bridge the modality gap between dialogues and images, we propose a dialogue-refined cross-modal alignment (DiaNA) framework, which leverages two adaptive attribute refiner modules to bottleneck the conversational and visual information for fine-grained cross-modal alignment. Moreover, we propose a dialogue-specific data augmentation strategy, random round retaining, to further enhance the model’s generalization ability across varying dialogue lengths. Extensive experiments demonstrate that DiaNA significantly outperforms existing TPR approaches, highlighting the effectiveness of conversational interactions for person retrieval.

Multimodal Emotion Recognition (MER) aims to predict human emotions by leveraging multiple modalities, such as vision, acoustics, and language. However, due to the heterogeneity of these modalities, MER faces two key challenges: modality balance dilemma and modality specialization disappearance. Existing methods often overlook the varying importance of modalities across samples in tackling the modality balance dilemma. Moreover, mainstream decoupling methods, while preserving modality-specific information, often neglect the predictive capability of unimodal data. To address these, we propose a novel model, Modality-Specific Enhanced Dynamic Emotion Experts (EMOE), consisting of: (1) Mixture of Modality Experts for dynamically adjusting modality importance based on sample features, and (2) Unimodal Distillation to retain single-modality predictive ability within fused features. EMOE enables adaptive fusion by learning a unique modality weight distribution for each sample, enhancing multi-modal predictions with single-modality predictions to balance invariant and specific features in emotion recognition. Experimental results on benchmark datasets show that EMOE achieves superior or comparable performance to state-of-the-art methods. Additionally, we extend EMOE to Multimodal Intent Recognition (MIR), further demonstrating its effectiveness and versatility.

Federated Graph Learning (FGL) has emerged as a solution to address real-world privacy concerns and data silos in graph learning, which relies on Graph Neural Networks (GNNs). Nevertheless, the homophily level discrepancies within the local graph data of clients, termed homophily heterogeneity, significantly degrade the generalizability of a global GNN. Existing research ignores this issue and suffers from unpromising collaboration. In this paper, we propose FedSPA, an effective framework that addresses homophily heterogeneity from the perspectives of homophily conflict and homophily bias. In the first place, the homophily conflict arises when training on inconsistent homophily levels across clients. Correspondingly, we propose Subgraph Feature Propagation Decoupling (SFPD), thereby achieving collaboration on unified homophily levels across clients. To further address homophily bias, we design Homophily Bias-Driven Aggregation (HBDA) which emphasizes clients with lower biases. It enables the adaptive adjustment of each client contribution to the global GNN based on its homophily bias. The superiority of FedSPA is validated through extensive experiments. The code is available at https://github.com/OakleyTan/FedSPA.

Pre-trained Vision-Language Models (VLMs) require Continual Learning (CL) to efficiently update their knowledge and adapt to various downstream tasks without retraining from scratch. However, for VLMs, in addition to the loss of knowledge previously learned from downstream tasks, pre-training knowledge is also corrupted during continual fine-tuning. This issue is exacerbated by the unavailability of original pre-training data, leaving VLM’s generalization ability degrading. In this paper, we propose GIFT, a novel continual fine-tuning approach that utilizes synthetic data to overcome catastrophic forgetting in VLMs. Taking advantage of recent advances in text-to-image synthesis, we employ a pre-trained diffusion model to recreate both pre-training and learned downstream task data. In this way, the VLM can revisit previous knowledge through distillation on matching diffusion-generated images and corresponding text prompts. Leveraging the broad distribution and high alignment between synthetic image-text pairs in VLM’s feature space, we propose a contrastive distillation loss along with an image-text alignment constraint. To further combat in-distribution overfitting and enhance distillation performance with limited amount of generated data, we incorporate adaptive weight consolidation, utilizing Fisher information from these synthetic image-text pairs and achieving a better stability-plasticity balance. Extensive experiments demonstrate that our method consistently outperforms previous state-of-the-art approaches across various settings.

Person re-identification (ReID) is the task of matching individuals across different camera views. Existing approaches typically employ neural networks to extract discriminative features, ranking gallery images based on their similarities to probe images. While effective, these methods are often enhanced through re-ranking, a post-processing step that refines initial retrieval results without requiring additional model training. However, current re-ranking methods mostly rely on k-nearest neighbor search to extract similar images that might have the same identity as the query, which is time-consuming with a high computation burden, limiting their applications in reality. We rethink the effect of the k-nearest neighbor search and introduce the Chebyshev’s Theorem-guided Graph Re-ranking (Cheb-GR) method, which adopts the adaptive neighbor search guided by Chebyshev’s Theorem over the k-nearest neighbor search for efficient neighbor selection. Our method leverages graph convolution operations to refine image features and achieve robust re-ranking, leading to enhanced retrieval performance. Furthermore, we provide a theoretical analysis based on Chebyshev’s Inequality to elucidate the factors contributing to the strong performance of the proposed method. Our method significantly reduces the computation costs while maintaining relatively strong performance. Through extensive experiments in both general and cross-domain settings, we demonstrate the effectiveness of Cheb-GR and its potential for real-world applications.

Understanding intention and emotion from social media poses unique challenges due to the inherent uncertainty in multimodal data, where posts often contain incomplete or missing modalities. While this uncertainty reflects real-world scenarios, it remains underexplored within the computer vision community, particularly in conjunction with the intrinsic relationship between emotion and intention. To address these challenges, we introduce the Multimodal IntentioN and Emotion Understanding in the Wild (MINE) dataset, comprising over 20,000 topic-specific social media posts with natural modality variations across text, image, video, and audio. MINE is distinctively constructed to capture both the uncertain nature of multimodal data and the implicit correlations between intentions and emotions, providing extensive annotations for both aspects. To tackle these scenarios, we propose the Bridging Emotion-Intention via Implicit Label Reasoning (BEAR) framework. BEAR consists of two key components: a BEIFormer that leverages emotion-intention correlations, and a Modality Asynchronous Prompt that handles modality uncertainty. Experiments show that BEAR outperforms existing methods in processing uncertain multimodal data while effectively mining emotion-intention relationships for social media content understanding.

Automatic and accurate vessel segmentation is crucial for disease diagnosis. Deep learning methods are widely used, but their promising results rely on accurately annotated data. Due to complex vessel morphology and low-contrast image, accurate vessel delineation poses a practical challenge, resulting in insufficient annotations, which is a prominent form of noisy labels. This paper proposes an Image-assisted Label Connective Completion method, which enhances label’s vessel information by images under the supervision of connectivity to address insufficient annotation issue. Specifically, we develop an Image-guided Vessel Enhancement module, which transmits structural information extracted from images based on label navigation to label space, promoting completion of missing annotated parts in original labels. In addition, a branch completion-connectivity loss is designed and introduced as an auxiliary supervision to prevent vessel branch disconnection during label completion. Experimental results on DRIVE, CHASE DB1 and DCA1 datasets demonstrate that our method outperforms existing noisy labels learning methods.

To advance real-world fashion image editing, we analyze existing two-stage pipelines(mask generation followed by diffusion-based editing)which overly prioritize generator optimization while neglecting mask controllability. This results in two critical limitations: I) poor user-defined flexibility (coarse-grained human masks restrict edits to predefined regions like upper torso; fine-grained clothes masks preserve poses but forbid style/length customization). II) weak pose robustness (mask generators fail due to articulated poses and miss rare regions like waist, while human parsers remain limited by predefined categories). To address these gaps, we propose Pose-Star, a framework that dynamically recomposes body structures (e.g., neck, chest, etc.) into anatomy-aware masks (e.g., chest-length) for user-defined edits. In Pose-Star, we calibrate diffusion-derived attention (Star tokens) via skeletal keypoints to enhance rare structure localization in complex poses, suppress noise through phase-aware analysis of attention dynamics (Convergence,Stabilization,Divergence) with threshold masking and sliding-window fusion, and refine edges via cross-self attention merging and Canny alignment. This work bridges controlled benchmarks and open-world demands, pioneering anatomy-aware, pose-robust editing and laying the foundation for industrial fashion image editing.

Federated Graph Learning is rapidly evolving as a privacy-preserving collaborative approach. However, backdoor attacks are increasingly undermining federated systems by injecting carefully designed triggers that lead to the model making incorrect predictions. Trigger structures and injection locations in Federated Graph Learning are more diverse, making traditional federated defense methods less effective. In our work, we propose an effective Federated Graph Backdoor Defense using Topological Graph Energy (FedTGE). At the local client level, it injects distribution knowledge into the local model, assigning low energy to benign samples and high energy to the constructed malicious substitutes, and selects benign clients through clustering. At the global server level, the energy elements uploaded by each client are treated as new nodes to construct a global energy graph for energy propagation, making the selected clients' energy elements more similar and further adjusting the aggregation weights. Our method can handle high data heterogeneity, does not require a validation dataset, and is effective under both small and large malicious proportions. Extensive results on various settings of federated graph scenarios under backdoor attacks validate the effectiveness of this approach.

Sketch-Based Person Re-Identification (Sketch ReID) aims to retrieve a person of interest across disjoint cameras using hand-drawn sketches as queries. A significant issue is the limited structural clues of sketch queries, which fail to convey high-level semantic retrieval intentions, such as colors and genders. Existing works typically combine sketches and texts for multi-modal retrieval, which inevitably introduces modality interference and relies heavily on expensive tri-modal datasets. In this paper, we propose, for the first time, an interactive and flexible sketch-based person retrieval framework that incorporates user feedback to refine the sketch person retrieval ranking without text training. A lightweight vision-to-text converting network is proposed to represent sketches with equivalent pseudo-word tokens, which aims to provide context assistance for interactive retrieval. Then, the sketch token can be seamlessly integrated with text feedback tokens within CLIP’s textual space for explicit sketch-text compositionality, thus achieving feedback-guided ranking refinement. Extensive experiments underscore the superiority of our InteractReID. Code will be available at https://github.com/littlexinyi/InteractReID.

Multimodal Large Language Model (MLLM) has demonstrated strong generalization capabilities across diverse distributions and tasks, largely due to extensive pre-training datasets. Fine-tuning MLLM has become a common practice to improve performance on specific downstream tasks. However, during fine-tuning, MLLM often faces the risk of forgetting knowledge acquired during pre-training, which can result in a decline in generalization abilities. To balance the trade-off between generalization and specialization, we propose measuring the parameter importance for both pre-trained and fine-tuning distributions, based on frozen pre-trained weight magnitude and accumulated fine-tuning gradient values. We further apply an importance-aware weight allocation strategy, selectively updating relatively important parameters for downstream tasks. We conduct empirical evaluations on both image captioning and visual question-answering tasks using various MLLM architectures. The comprehensive experimental analysis demonstrates the effectiveness of the proposed solution, highlighting the efficiency of the crucial modules in enhancing downstream specialization performance while mitigating generalization degradation in MLLM Fine-Tuning.

Fine-tuning Multimodal Large Language Models (MLLMs) in multi-task learning scenarios has emerged as an effective strategy for achieving cross-domain specialization. However, multi-task fine-tuning frequently induces performance degradation on open-response datasets. We posit that free-form answer generation primarily depends on language priors, and strengthening the integration of visual behavioral cues is critical for enhancing prediction robustness. In this work, we propose Noise Resilient Confidence Alignment to address the challenge of open-response overfitting during multi-task fine-tuning. Our approach prioritizes maintaining consistent prediction patterns in MLLMs across varying visual input qualities. To achieve this, we employ Gaussian perturbations to synthesize distorted visual inputs and enforce token prediction confidence alignment towards the normal visual branch. By explicitly linking confidence calibration to visual robustness, this method reduces over-reliance on language priors. We conduct extensive empirical evaluations across diverse multi-task downstream settings via popular MLLM architectures. The comprehensive experiment demonstrates the effectiveness of our method, showcasing its ability to alleviate open-response overfitting while maintaining satisfying multi-task fine-tuning performance.

Multimodal large language models (MLLMs) have achieved impressive progress in tasks such as visual question answering and visual understanding, but they still face significant challenges in emotional reasoning. Current methods to enhance emotional understanding typically rely on fine-tuning or manual annotations, which are resource-intensive and limit scalability. In this work, we focus on improving the ability of MLLMs to capture emotions during the inference phase. Specifically, MLLMs encounter two main issues: they struggle to distinguish between semantically similar emotions, leading to misclassification, and they are overwhelmed by redundant or irrelevant visual information, which distracts from key emotional cues. To address these, we propose Sharpening Emotion Perception in MLLMs (SEPM), which incorporates a Confidence-Guided Coarse-to-Fine Inference framework to refine emotion classification by guiding the model through simpler tasks. Additionally, SEPM employs Focus-on-Emotion Visual Augmentation to reduce visual redundancy by directing the attention of models to relevant emotional cues in images. Experimental results demonstrate that SEPM significantly improves MLLM performance on emotion-related tasks, providing a resource-efficient and scalable solution for emotion recognition.

Federated Prompt Learning (FPL) adapts pre-trained Vision-Language Models (VLMs) to federated learning through prompt tuning, leveraging their transferable representations and strong generalization capabilities. Traditional methods often require uniform prompt lengths for federated aggregation, limiting adaptability to clients with diverse prompt lengths and distribution biases. In this paper, we propose **Fed**erated **P**rompt Learning for **H**eterogeneous Client **A**daptation (FedPHA), a novel framework that combines a fixed-length global prompt for efficient aggregation with local prompts of varying lengths to capture client-specific data characteristics. Additionally, FedPHA designs Singular Value Decomposition (SVD) based projection and bidirectional alignment to disentangle global conflicts arising from client heterogeneity, ensuring that personalized client tasks effectively utilize non-harmful global knowledge. This approach ensures that global knowledge improves model generalization while local knowledge preserves local optimization. Experimental results validate the effectiveness of FedPHA in achieving a balance between global and personalized knowledge in federated learning scenarios.

Federated Learning (FL) enables collaborative training with privacy preservation but is vulnerable to backdoor attacks, where malicious clients degrade model performance on targeted inputs. These attacks exploit FL decentralized nature, while existing defenses, based on isolated behaviors and fixed rules, can be bypassed by adaptive attackers. To address these limitations, we propose **SPMC**, a marginal collaboration defense mechanism that leverages intrinsic consistency across clients to estimate inter-client marginal contributions. This allows the system to dynamically reduce the influence of clients whose behavior deviates from the collaborative norm, thus maintaining robustness even as the number of attackers changes. In addition to overcoming proxy-dependent purification's weaknesses, we introduce a self-purification process that locally adjusts suspicious gradients. By aligning them with margin-based model updates, we mitigate the effect of local poisoning. Together, these two modules significantly improve the adaptability and resilience of FL systems, both at the client and server levels. Experimental results on a variety of classification benchmarks demonstrate that SPMC achieves strong defense performance against sophisticated backdoor attacks without sacrificing accuracy on benign tasks. The code is posted at: https://github.com/WenddHe0119/SPMC.

Federated learning provides an efficient privacy-preserving distributed training framework for large language models, addressing the growing scarcity of publicly available training data while enabling the utilization of private datasets. While integrating large language model fine-tuning with federated learning emerges as a promising research direction, researchers pay limited attention to non-IID instruction-following scenarios. Our key insight is decomposing client updates into consensus and divergence components, enabling the model to maintain core capabilities while adapting to domain-specific knowledge. We propose a novel federated learning framework called **FedICU** (Splitting with **I**mportan**C**e-aware **U**pdating for Heterogeneous **Fed**erated Learning with Large Language Models), which introduces an aggregation mechanism that dynamically balances these components based on their contribution to global model performance, while implementing an importance-aware parameter updating strategy to prevent catastrophic forgetting and domain overfitting. Extensive experiments across diverse domains demonstrate that FedICU significantly outperforms existing federated learning approaches in terms of both generalization performance and domain adaptation. Our code is available at https://github.com/liaosunny123/FedICU.

Federated Graph Learning (FGL) proposes an effective approach to collaboratively training Graph Neural Networks (GNNs) while maintaining privacy. Nevertheless, communication efficiency becomes a critical bottleneck in environments with limited resources. In this context, one-shot FGL emerges as a promising solution by restricting communication to a single round. However, prevailing FGL methods face two key challenges in the one-shot setting: 1) They heavily rely on gradual personalized optimization over multiple rounds, undermining the capability of the global model to efficiently generalize across diverse graph structures. 2) They are prone to overfitting to local data distributions due to extreme structural bias, leading to catastrophic forgetting. To address these issues, we introduce **GHOST**, an innovative one-shot FGL framework. In GHOST, we establish a proxy model for each client to leverage diverse local knowledge and integrate it to train the global model. During training, we identify and consolidate parameters essential for capturing topological knowledge, thereby mitigating catastrophic forgetting. Extensive experiments on real-world tasks demonstrate the superiority and generalization capability of GHOST. The code is available at https://github.com/JiaruQian/GHOST.

Generative replay (GR) has been extensively validated in continual learning as a mechanism to synthesize data and replay past knowledge to mitigate forgetting. By leveraging synthetic rather than real data for the replay, GR has been adopted in some federated continual learning (FCL) approaches to ensure the privacy of client-side data. While existing GR-based FCL approaches have introduced improvements, none of their enhancements specifically take into account the unique characteristics of federated learning settings. Beyond privacy constraints, what other fundamental aspects of federated learning should be explored in the context of FCL? In this work, we explore the potential benefits that come from emphasizing the role of clients throughout the process. We begin by highlighting two key observations: (a) Client Expertise Superiority, where clients, rather than the server, act as domain experts, and (b) Client Forgetting Variance, where heterogeneous data distributions across clients lead to varying levels of forgetting. Building on these insights, we propose CAN (Clients As Navigators), highlighting the pivotal role of clients in both data synthesis and data replay. Extensive evaluations demonstrate that this client-centric approach achieves state-of-the-art performance. Notably, it requires a smaller buffer size, reducing storage overhead and enhancing computational efficiency.

Federated Graph Learning (FGL) has gained significant attention as a privacy-preserving approach to collaborative learning, but the computational demands increase substantially as datasets grow and Graph Neural Network (GNN) layers deepen. To address these challenges, we propose $\textbf{EAGLES}$, a unified sparsification framework. EAGLES applies client-consensus parameter sparsification to generate multiple unbiased subnetworks at varying sparsity levels, reducing the need for iterative adjustments and mitigating performance degradation. In the graph structure domain, we introduced a dual-expert approach: a $\textit{graph sparsification expert}$ uses multi-criteria node-level sparsification, and a $\textit{graph synergy expert}$ integrates contextual node information to produce optimal sparse subgraphs. Furthermore, the framework introduces a novel distance metric that leverages node contextual information to measure structural similarity among clients, fostering effective knowledge sharing. We also introduce the $\textbf{Harmony Sparsification Principle}$, EAGLES balances model performance with lightweight graph and model structures. Extensive experiments demonstrate its superiority, achieving competitive performance on various datasets, such as reducing training FLOPS by 82\% $\downarrow$ and communication costs by 80\% $\downarrow$ on the ogbn-proteins dataset, while maintaining high performance.

Federated Graph Learning (FGL) combines the privacy-preserving capabilities of Federated Learning (FL) with the strong graph modeling capability of Graph Neural Networks (GNNs). Current research addresses subgraph-FL from the structural perspective, neglecting the propagation of graph signals on the spatial and spectral domains of the structure. From a spatial perspective, subgraph-FL introduces edge disconnections between clients, leading to disruptions in label signals and a degradation in the semantic knowledge of the global GNN. From a spectral perspective, spectral heterogeneity causes inconsistencies in signal frequencies across subgraphs, which makes local GNNs overfit the local signal propagation schemes. As a result, spectral client drift occurs, undermining global generalizability. To tackle the challenges, we propose a global knowledge repository to mitigate the challenge of poor semantic knowledge caused by label signal disruption. Furthermore, we design a frequency alignment to address spectral client drift. The combination of Spatial and Spectral strategies forms our framework $S^2$FGL. Extensive experiments on multiple datasets demonstrate the superiority of $S^2$FGL. The code is available at https://github.com/Wonder7racer/S2FGL.git.

Federated Parameter-Efficient Fine-Tuning aims to adapt Vision-Language Models for downstream tasks in distributed environments. However, data heterogeneity across participants hinders collaborative effectiveness, necessitating personalized adaptation to cover distinct data distributions. Current personalized methods suffer from two limitations. 1) Textual Property Loss: Existing methods facilitate the collaboration between decoupled prompts at the feature level, which potentially undermines the textual properties of the prompts. 2) Visual Feature Diversity: The diversity of visual features makes it challenging to leverage naive image features directly for image-text alignment in downstream tasks. In this work, we propose Federated Disentangled Tuning with Textual Prior Decoupling and Visual Dynamic Adaptation (FedDDA) to overcome the above limitations. Specifically, we encourage decoupling prompts in a way that maximizes the efficacy of prior knowledge, which is essential for maintaining a coherent linguistic context. Furthermore, we design a visual adaption model to reshape visual space to optimally align with the textual space. Extensive experiments on various image classification tasks show the effectiveness of our work in addressing data heterogeneity. The codes are released at https://github.com/MoratalYang/FedDDA.

Accurate vessel segmentation is essential for diagnosing and managing vascular and ophthalmic diseases. Traditional learning-based vessel segmentation methods heavily rely on high-quality, pixel-level annotated datasets. However, segmentation performance suffers significantly when applied in federated learning settings due to vessel morphology inconsistency and vessel-background imbalance. The former limits the ability of models to capture fine-grained vessels, while the latter overemphasizes background pixels and biases the model towards them. To address these challenges, we propose a novel method named Federated Vessel-Aware Calibration (FVAC), which leverages global uncertainty to provide differentiated guidance for clients, focusing on pixels of various morphologies that are difficult to distinguish. Furthermore, we introduce a foreground-background decoupling alignment strategy that utilizes more stable and balanced global features to mitigate semantic drift caused by vessel-background imbalance in local clients. Comprehensive experiments confirm the effectiveness of our method

In real applications, person re-identification (ReID) expects to retrieve the target person at any time, including both daytime and nighttime, ranging from short-term to long-term. However, existing ReID tasks and datasets cannot meet this requirement, as they are constrained by available time and only provide training and evaluation for specific scenarios. Therefore, we investigate a new task called Anytime Person Re-identification (AT-ReID), which aims to achieve effective retrieval in multiple scenarios based on variations in time. To address the AT-ReID problem, we collect the first large-scale dataset, AT-USTC, which contains 135k images of individuals wearing multiple clothes captured by RGB and IR cameras. Our data collection spans over an entire year and 270 volunteers were photographed on average 29.1 times across different dates or scenes, 4-15 times more than current datasets, providing conditions for follow-up investigations in AT-ReID. Further, to tackle the new challenge of multi-scenario retrieval, we propose a unified model named Uni-AT, which comprises a multi-scenario ReID (MS-ReID) framework for scenario-specific features learning, a Mixture-of-Attribute-Experts (MoAE) module to alleviate inter-scenario interference, and a Hierarchical Dynamic Weighting (HDW) strategy to ensure balanced training across all scenarios. Extensive experiments show that our model leads to satisfactory results and exhibits excellent generalization to all scenarios.

Lifelong Person Re-identification (LReID) is essential in dynamic camera networks, which continually adapts to new environments while preserving previously acquired knowledge. Existing LReID techniques often preserve samples from past datasets to maintain old knowledge, potentially leading to privacy risks. While prototype-based methods offer privacy advantages, current approaches primarily focus on adjusting classifiers for image classification tasks, neglecting representation biases between old and new identities in person re-identification. This study introduces a novel Prototype-guided Knowledge Propagation (PKP) method, which mitigates discrepancies in similar identity images between old and new tasks by guiding prototype construction through triplet loss constraints. Additionally, to address disparities between prototypes and the updated feature extractor, an Adaptive Parameter Evolution (APE) strategy is proposed. APE optimizes the integration of the old and new models by assessing the importance of the new tasks, dynamically selecting the most pertinent parameters for updates according to their contribution to the current task. Extensive experiments on the LReID benchmark demonstrate that our approach surpasses state-of-the-art prototype-based LReID methods in terms of mAP and rank-1 accuracy. Code is available at https://github.com/joyner-7/IJCAI2025-PKA.

The Vision Language Model (VLM) excels in aligning vision and language representations, and prompt learning has emerged as a key technique for adapting such models to downstream tasks. However, the application of prompt learning with VLM in federated learning (FL) scenarios remains underexplored. This paper systematically investigates the behavioral differences between language prompt learning (LPT) and vision prompt learning (VPT) under data heterogeneity challenges, including label skew and domain shift. We conduct extensive experiments to evaluate the impact of various FL and prompt configurations, such as client scale, aggregation strategies, and prompt length, to assess the robustness of Federated Prompt Learning (FPL). Furthermore, we explore strategies for enhancing prompt learning in complex scenarios where label skew and domain shift coexist, including leveraging both prompt types when computational resources allow. Our findings offer practical insights into optimizing prompt learning in federated settings, contributing to the broader deployment of VLMs in privacy-preserving environments.

Person re-identification (Re-ID) models have achieved remarkable advancements with the advent of deep learning. However, their performance often degrades in diverse scenarios, such as variations in viewing angles, lighting conditions, and environmental changes. These limitations arise from the difficulty in generalizing across multiple factors, including environments and subject appearances. Multimodal Large Language Models (MLLMs) offer a promising alternative to address these challenges by leveraging generalized knowledge, as demonstrated in biometric tasks like face and iris recognition. This study explores the Re-ID capabilities of MLLMs by comparatively evaluating six representative MLLMs on the most challenging scenarios, including angle variation, illumination differences, clothing changes, image corruption, and visually fine-grained scenarios in Re-ID. We find that GPT-4o outperforms other MLLMs in handling angle variation, illumination differences, corruption resistance, and fine-grained detail disturbances, demonstrating high accuracy and robustness in challenging Re-ID scenarios. However, further optimization is required for robustness against illumination variation, corruption handling, and fine-grained identification across all tested MLLMs. Additionally, the Re-ID performance of MLLMs can be improved by applying several prompt templates. Our research suggests potential directions for integrating MLLMs into Re-ID systems to enhance performance and robustness, underscoring their promising potential in this field.

Federated Learning enables collaborative training across decentralized edge devices while preserving data privacy. However, fine-tuning large-scale pre-trained models in federated learning is hampered by substantial communication overhead and client resource limitations. Parameter-efficient fine-tuning methods like Low-Rank Adaptation (LoRA) reduce resource demands but suffer from aggregation discrepancies and heightened vulnerability to label noise, particularly in heterogeneous federated settings. In this paper, we introduce RFedLR, a robust federated PEFT framework designed to overcome these challenges. RFedLR integrates two key components: (1) Sensitivity-aware robust tuning, which identifies and selectively updates noise-sensitive parameters to bolster local robustness against label noise, and (2) Adaptive federated LoRA aggregation, which dynamically weights and aggregates LoRA updates based on their importance and stability to minimize bias and noise propagation. Comprehensive experimental validation shows RFedLR outperforms existing methods, achieving superior accuracy and robustness in noisy federated scenarios. Our code is available at: https://github.com/FangXiuwen/RFedLR

Federated Learning is a promising technique that enables collaborative machine learning while preserving participant privacy. With respect to multi-party collaboration, achieving performance fairness acts as a critical challenge in federated systems. Existing explorations mainly focus on considering all parameter-wise fairness and consistently protecting weak clients to achieve performance fairness in federation. However, these approaches neglect two critical issues. 1) Parameter Redundancy: Redundant parameters that are unnecessary for fairness training may conflict with critical parameters update, thereby leading to performance degradation. 2) Persistent Protection: Current fairness mechanisms persistently enhance weak clients throughout the entire training cycle, hindering global optimization and causing lower performance alongside unfairness. To address these, we propose a strategy with two key components: First, parameter adjustment with mask and rescale which discarding redundant parameter and highlight critical ones, preserving key parameter updates and decrease conflict. Second, we observe that the federated training process exhibits distinct characteristics across different phases. We propose a dynamic aggregation strategy that adaptively weights clients based on local update directions and performance variations. Empirical results on single-domain and cross-domain scenarios demonstrate the effectiveness of the proposed solution and the efficiency of crucial modules. The code is available at https://github.com/guankaiqi/FedPW.

Multimodal Large Language Models (MLLMs) are increasingly deployed in fine-tuning-as-a-service (FTaaS) settings, where user-submitted datasets adapt general-purpose models to downstream tasks. This flexibility, however, introduces serious security risks, as malicious fine-tuning can implant backdoors into MLLMs with minimal effort. In this paper, we observe that backdoor triggers systematically disrupt cross-modal processing by causing abnormal attention concentration on non-semantic regions—a phenomenon we term **attention collapse**. Based on this insight, we propose **Believe Your Eyes (BYE)**, a data filtering framework that leverages attention entropy patterns as self-supervised signals to identify and filter backdoor samples. BYE operates via a three-stage pipeline: (1) extracting attention maps using the fine-tuned model, (2) computing entropy scores and profiling sensitive layers via bimodal separation, and (3) performing unsupervised clustering to remove suspicious samples. Unlike prior defenses, BYE equires no clean supervision, auxiliary labels, or model modifications. Extensive experiments across various datasets, models, and diverse trigger types validate BYE's effectiveness: it achieves near-zero attack success rates while maintaining clean-task performance, offering a robust and generalizable solution against backdoor threats in MLLMs.

Vertical Federated Learning (VFL) has emerged as a critical privacy-preserving learning paradigm, enabling collaborative model training by leveraging distributed features across clients. However, due to privacy concerns, there are few publicly available real-world datasets for evaluating VFL methods, which poses significant challenges to related research. To bridge this gap, we propose MARS-VFL, a unified benchmark for realistic VFL evaluation. It integrates data from practical applications involving collaboration across different features, maintaining compatibility with the VFL setting. Based on this, we standardize the evaluation of VFL methods from the mainstream aspects of efficiency, robustness, and security. We conduct comprehensive experiments to assess different VFL approaches, providing references for unified evaluation. Furthermore, we are the first to unify the evaluation of robustness challenges in VFL and introduce a new method for addressing robustness challenges, establishing standard baselines for future research.

Federated Graph Learning (FGL) has been shown to be particularly effective in enabling collaborative training of Graph Neural Networks (GNNs) in decentralized settings. Model-heterogeneous FGL further enhances practical applicability by accommodating client preferences for diverse model architectures. However, existing model-heterogeneous approaches primarily target Euclidean data and fail to account for a crucial aspect of graph-structured data: topological relationships. To address this limitation, we propose **TRUST**, a novel knowledge distillation-based **model-heterogeneous FGL** framework. Specifically, we propose Progressive Curriculum Node Scheduler to progressively introduce challenging nodes based on learning difficulty. In Adaptive Curriculum Distillation Modulator, we propose an adaptive temperature modulator that dynamically adjusts knowledge distillation temperature to accommodate varying client capabilities and graph complexity. Moreover, we leverage Wasserstein‑Driven Affinity Distillation to enable models to capture cross-class structural relationships through optimal transport. Extensive experiments on multiple graph benchmarks and model-heterogeneous settings show that **TRUST** outperforms existing methods, achieving an average 3.6\% $\uparrow$ performance gain, particularly under moderate heterogeneity conditions. The code is available for anonymous access at https://anonymous.4open.science/r/TRUST-NeurIPS2025.

Federated Graph Learning (FGL) offers a promising framework for collaboratively training Graph Neural Networks (GNNs) while preserving data privacy. In resource-constrained environments, One-shot Federated Learning (OFL) emerges as an effective solution by limiting communication to a single round. Current OFL approaches employing generative models have attracted considerable attention; however, they face unresolved challenges: these methods are primarily designed for traditional image data and fail to capture the fine-grained structural information of local graph data. Consequently, they struggle to integrate the intricate correlations necessary and transfer subtle structural insights from each client to the global model. To address these issues, we introduce **OASIS**, an innovative one-shot FGL framework. In OASIS, we propose a Synergy Graph Synthesizer designed to generate informative synthetic graphs and introduce a Topological Codebook to construct a structural latent space. Moreover, we propose the Wasserstein-Enhanced Semantic Affinity Distillation (WESAD) to incorporate rich inter-class relationships and the Wasserstein-Driven Structural Relation Distillation (WDSRD) to facilitate the effective transfer of structural knowledge from the Topological Codebook. Extensive experiments on real-world tasks demonstrate the superior performance and generalization capability of OASIS. The code is available for anonymous access at https://anonymous.4open.science/r/OASIS-NeurIPS25.

Graph neural networks (GNNs) have achieved remarkable success in various domains but typically rely on centralized, static graphs, which limits their applicability in distributed, evolving environments. To address this limitation, we define the task of Federated Continual Graph Learning (FCGL), a paradigm for incremental learning on dynamic graphs distributed across decentralized clients. Existing methods, however, neither preserve graph topology during task transitions nor mitigate parameter conflicts in server‐side aggregation. To overcome these challenges, we introduce **MOTION**, a generalizable FCGL framework that integrates two complementary modules: the Graph Topology‐preserving Multi‐Sculpt Coarsening (G‐TMSC) module, which maintains the structural integrity of past graphs through a multi‐expert, similarity‐guided fusion process, and the Graph‐Aware Evolving Parameter Adaptive Engine (G‐EPAE) module, which refines global model updates by leveraging a topology‐sensitive compatibility matrix. Extensive experiments on real‐world datasets show that our approach improves average accuracy (AA) by an average of 30\% $\uparrow$ over the FedAvg baseline across five datasets while maintaining a negative $\downarrow$ average forgetting (AF) rate, significantly enhancing generalization and robustness under FCGL settings. The code is available for anonymous access at https://anonymous.4open.science/r/MOTION.

The application of AI in psychiatric diagnosis faces significant challenges, including the subjective nature of mental health assessments, symptom overlap across disorders, and privacy constraints limiting data availability. To address these issues, we present MoodAngels, the first specialized multi-agent framework for mood disorder diagnosis. Our approach combines granular-scale analysis of clinical assessments with a structured verification process, enabling more accurate interpretation of complex psychiatric data. Complementing this framework, we introduce MoodSyn, an open-source dataset of 1,173 synthetic psychiatric cases that preserves clinical validity while ensuring patient privacy. Experimental results demonstrate that MoodAngels outperforms conventional methods, with our baseline agent achieving 12.3\% higher accuracy than GPT-4o on real-world cases, and our full multi-agent system delivering further improvements. Together, these contributions provide both an advanced diagnostic tool and a critical research resource for computational psychiatry, bridging important gaps in AI-assisted mental health assessment.

Multimodal Intention Recognition (MIR) plays a critical role in applications such as intelligent assistants, service robots, and autonomous systems. However, in real-world settings, different modalities often vary significantly in informativeness, reliability, and noise levels. This leads to modality imbalance, where models tend to over-rely on dominant modalities, thereby limiting generalization and robustness. While existing methods attempt to alleviate this issue at either the sample or model level, most overlook its multi-level nature. To address this, we propose Adaptive Re-calibration Learning (ARL), a novel dual-path framework that models modality importance from both sample-wise and structural perspectives. ARL incorporates two key mechanisms: Contribution-Inverse Sample Calibration (CISC), which dynamically masks overly dominant modalities at the sample level to encourage attention to underutilized ones; and Weighted Encoder Calibration (WEC), which adjusts encoder weights based on global modality contributions to prevent overfitting. Experimental results on multiple MIR benchmarks demonstrate that ARL significantly outperforms existing methods in both accuracy and robustness, particularly under noisy or modality-degraded conditions.

Federated Learning (FL) has demonstrated a promising future in privacy-friendly collaboration but it faces the data heterogeneity problem. Knowledge Distillation (KD) can serve as an effective method to address this issue. However, challenges arise from the unreliability of existing distillation methods in multi-domain scenarios. Prevalent distillation solutions primarily aim to fit the distributions of the global model directly by minimizing forward Kullback-Leibler divergence (KLD). This results in significant bias when the outputs of the global model are multi-peaked, which indicates the unreliability of the distillation pathway. Meanwhile, cross-domain update conflicts can notably reduce the accuracy of the global model (teacher model) in certain domains, reflecting the unreliability of the teacher model in these domains. In this work, we propose DKDR (Dynamic Knowledge Distillation for Reliability in Federated Learning), which dynamically assigns weights to forward and reverse KLD based on knowledge discrepancies. This enables clients to fit the outputs from the teacher precisely. Moreover, we use knowledge decoupling to identify domain experts, thus clients can acquire reliable domain knowledge from experts. Empirical results from single-domain and multi-domain image classification tasks demonstrate the effectiveness of the proposed method and the efficiency of its key modules. The code is available at https://github.com/YueyangYuan/DKDR.

With the exponential growth of video content, aiming at localizing relevant video moments based on natural language queries, video moment retrieval (VMR) has gained significant attention. Existing weakly supervised VMR methods focus on designing various feature modeling and modal interaction modules to alleviate the reliance on precise temporal annotations. However, these methods have poor generalization capabilities on compositional queries with novel syntactic structures or vocabulary in real-world scenarios. To this end, we propose a new task: weakly supervised compositional moment retrieval (WSCMR). This task trains models using only video-query pairs without precise temporal annotations, while enabling generalization to complex compositional queries. Furthermore, a proposal-centric network (PC-Net) is proposed to tackle this challenging task. First, video and query features are extracted through frozen feature extractors, followed by modality interaction to obtain multimodal features. Second, to handle compositional queries with explicit temporal associations, a dual-granularity proposal generator decodes multimodal global and frame-level features to obtain query-relevant proposal boundaries with fine-grained temporal perception. Third, to improve the discrimination of proposal features, a proposal feature aggregator is constructed to conduct semantic alignment of frames and queries, and employ a learnable peak-aware Gaussian distributor to fit the frame weights within the proposals to derive proposal features from the video frame features. Finally, the proposal quality is assessed based on the results of reconstructing the masked query using the obtained proposal features. To further enhance the model's ability to capture semantic associations between proposals and queries, a quality margin regularizer is constructed to dynamically stratify proposals into high and low query-relevance subsets and enhance the association between queries and common elements within proposals, and suppress spurious correlations via inter-subset contrastive learning. Notably, PC-Net achieves superior performance with 54\% fewer parameters than prior works by parameter-efficient design. Experiments on Charades-CG and ActivityNet-CG demonstrate PC-Net’s ability to generalize across diverse compositional queries. Code is available at https://github.com/mingyao1120/PC-Net.

In object re-identification (ReID) task, both cross-modal and multi-modal retrieval methods have achieved notable progress. However, existing approaches are designed for specific modality and category (person or vehicle) retrieval task, lacking generalizability to others. Acquiring multiple task-specific models would result in wasteful allocation of both training and deployment resources. To address the practical requirements for unified retrieval, we introduce Multi-Modal and Multi-Task object ReID ($\rm {M^3T}$-ReID). The $\rm {M^3T}$-ReID task aims to utilize a unified model to simultaneously achieve retrieval tasks across different modalities and different categories. Specifically, to tackle the challenges of modality distibution divergence and category semantics discrepancy posed in $\rm {M^3T}$-ReID, we design a novel Unbiased Prototype Consistency Learning (UPCL) framework, which consists of two main modules: Unbiased Prototypes-guided Modality Enhancement (UPME) and Cluster Prototype Consistency Regularization (CPCR). UPME leverages modality-unbiased prototypes to simultaneously enhance cross-modal shared features and multi-modal fused features. Additionally, CPCR regulates discriminative semantics learning with category-consistent information through prototypes clustering. Under the collaborative operation of these two modules, our model can simultaneously learn robust cross-modal shared feature and multi-modal fused feature spaces, while also exhibiting strong category-discriminative capabilities. Extensive experiments on multi-modal datasets RGBNT201 and RGBNT100 demonstrates our UPCL framework showcasing exceptional performance for $\rm {M^3T}$-ReID. The code is available at https://github.com/ZhouZhongao/UPCL.

This paper studies a challenging robust federated learning task with model heterogeneous and data corrupted clients, where the clients have different local model structures. Data corruption is unavoidable due to factors, such as random noise, compression artifacts, or environmental conditions in real-world deployment, drastically crippling the entire federated system. To address these issues, this paper introduces a novel Robust Asymmetric Heterogeneous Federated Learning (RAHFL) framework. We propose a Diversity-enhanced supervised Contrastive Learning technique to enhance the resilience and adaptability of local models on various data corruption patterns. Its basic idea is to utilize complex augmented samples obtained by the mixed-data augmentation strategy for supervised contrastive learning, thereby enhancing the ability of the model to learn robust and diverse feature representations. Furthermore, we design an Asymmetric Heterogeneous Federated Learning strategy to resist corrupt feedback from external clients. The strategy allows clients to perform selective one-way learning during collaborative learning phase, enabling clients to refrain from incorporating lower-quality information from less robust or underperforming collaborators. Extensive experimental results demonstrate the effectiveness and robustness of our approach in diverse, challenging federated learning environments.

In unsupervised meta-learning, the clustering-based pseudo-labeling approach is an attractive framework, since it is model-agnostic, allowing it to synergize with supervised algorithms to learn from unlabeled data. However, the pseudo-labels suffer from clustering noise and semantic chaos problems, further impacting the effectiveness of meta-learning. In this paper, we analyze and optimize the pseudo-labeling process, including encoding and clustering, aiming to generate semantic-like pseudo-labels to narrow the gap between unsupervised and supervised meta-learning. First, during the encoding, we observe that the embedding space of existing methods lacks clustering-friendly properties, which is the primary reason for clustering noise. To address this issue, we minimize the inter-to-intra-class similarity ratio to generate clustering-friendly embedding features and validate our approach through comprehensive experiments. Then, during the clustering, we find that the semantic quality of pseudo-labels is not adequately controlled, resulting in semantic chaos of pseudo-labels. We propose a semantic-stability index to measure the semantic quality of pseudo-labels quantitatively. Based on this index, we propose the Semantic-aware Pseudo-label Reassignment mechanism to generate semantic-like pseudo-labels for all samples. Our approach is model-agnostic and can easily be integrated into existing supervised methods. To demonstrate its generalization ability, we integrate it into two representative algorithms: MAML and EP. The results on three main few-shot benchmarks clearly show that the proposed method achieves significant improvement compared to state-of-the-art models. Notably, our approach also outperforms the corresponding supervised method in three tasks.

Visible-infrared person re-identification (VI-ReID) is a challenging cross-modality retrieval task due to the large modality gap. While numerous efforts have been devoted to the supervised setting with a large amount of labeled cross-modality correspondences, few studies have tried to mitigate the modality gap by mining cross-modality correspondences in an unsupervised manner. However, existing works failed to capture the intrinsic relations among samples across two modalities, resulting in limited performance outcomes. In this paper, we propose a novel Progressive Graph Matching (PGM) approach to globally model the cross-modality relationships and instance-level affinities. PGM formulates cross-modality correspondence mining as a graph matching procedure, aiming to integrate global information by minimizing global matching costs. Considering that samples in wrong clusters cannot find reliable cross-modality correspondences by PGM, we further introduce a robust Dual-Level Matching (DLM) mechanism, combining the cluster-level PGM and Nearest Instance-Cluster Searching (NICS) with instance-level affinity optimization. Additionally, we design an Outlier Filter Strategy (OFS) to filter out unreliable cross-modality correspondences based on the dual-level relation constraints. To mitigate false accumulation in cross-modal correspondence learning, an Alternate Cross Contrastive Learning (ACCL) module is proposed to alternately adjust the dominated matching, i.e., visible-to-infrared or infrared-to-visible matching. Empirical results demonstrate the superiority of our unsupervised solution, achieving comparable performance with supervised counterparts.

Conversational Emotion Recognition (CER) significantly benefits from the integration of multiple modalities. However, real-world scenarios are often plagued by hardware malfunctions and network failures that lead to missing modalities and incomplete emotional representations. Existing methods primarily focus on modeling inter-modal relationships within isolated utterances to generate missing data, so they do not adequately capture conversational context and dynamic emotional evolution. In conversations, emotional expressions are inherently context-dependent and dynamic. Insufficient modeling of these properties can substantially degrade performance. To address these challenges, we propose Contextual Graph Reconstruction and Emotional Variation Learning (CGR-EVL). Our approach flexibly constructs an utterance graph with diverse coverage based on speaker activity and modality availability, thereby capturing more contextual information. To enhance the handling of missing modalities, we further integrate temporal relations through a relational graph convolutional network to reconstruct missing features and introduce a speaker emotion-aware constraint to ensure emotional coherence. Additionally, we propose the concept of emotional entropy to quantify variation patterns and develop a novel loss function that aligns predicted and actual variations. Experiments on the IEMOCAP and MELD datasets show that CGR-EVL outperforms state-of-the-art methods, particularly under conditions of incomplete modalities.

As an intermediate task in computer vision, multiple pedestrian tracking (MPT) aiming at tracking the pedestrians from a given video, has attracted attention due to its potential academic and commercial value. However, pedestrians commonly suffer from occlusion due to diverse and complex scenarios, which increases the challenge of this task. This survey provides comprehensive review in terms of occlusion scenarios encountered during MPT, and investigates the model robustness of the existing methods in this scenarios. Firstly, this survey introduces the various and states of occlusion. Secondly, the related occlusion datasets are introduced. Subsequently, we categorize existing occlusion handling methods according to the tracking process and detail their pros and cons. In addition, occlusion handling precision (OHP) metric is proposed to evaluate the ability of a tracker in handling occlusion in this survey. Moreover, comprehensive analyzes and discussions in several public datasets are provided to verify the effectiveness of these methods. Finally, the existing issues and future directions for occlusion handling methods are discussed. In doing so, this work serves as a foundation for future research by providing researchers with information about the occlusion handling method of MPT.

UAV surveillance offers a unique aerial perspective, enabling the monitoring of large areas and capturing targets from angles that fixed ground cameras cannot achieve. UAV-based object re-identification (ReID) differs from the extensively studied city camera scenarios, as it involves identifying specific objects in aerial images captured from a dynamic bird’s-eye view. The challenge lies in the significant variation in object perspectives and the often uncertain rotational changes captured by UAVs. Existing ReID methods designed for city cameras struggle to adapt to these rotational variations. To address these challenges, we propose a Transformer-based learnable rotation generalization enhancement method specifically for UAV-based ReID. To improve the model’s adaptability to uncertain rotational changes, we introduce a learnable feature-level rotation simulation technique that generates multiple rotated features. Building on this, we design a rotation diversification loss to decorrelate different rotated features, ensuring a rich feature representation. Additionally, to mitigate the negative effects of image-level rotation augmentation, we propose instance-level and distribution-level rotation invariance regularization. This approach establishes explicit associations between images and their rotated counterparts, facilitating the learning of visually consistent rotation-invariant features. Instance-level constraints ensure that detailed features remain consistent during rotation, while distribution-level constraints maintain the model’s semantic understanding. Notably, our method demonstrates strong versatility, covering a wide range of objects, including persons, vehicles, and various animals. Evaluations on multiple UAV-collected person and vehicle ReID datasets, as well as several animal datasets, consistently show outstanding performance, underscoring its robustness and adaptability to the unique challenges posed by UAV-based ReID.

With the development of blockchain technology, Blockchain-based Crowdsensed Data Trading (BCDT) has emerged as an attractive data exchange paradigm. Although it addresses security issues in data transactions, most recent research primarily focuses on offline scenarios, overlooking the critical importance of enabling real-time online data trading, where it suffers from dynamic worker participation and potential malicious attacks. In this paper, we propose a Blockchain-based Secure and Fair Online Incentive Mechanism (BSFOIM), which primarily incorporates a smart contract called BSFOIMToken, designed to function in online scenarios. In particular, we first introduce a multi-stage auction combined with a time discount factor in BSFOIM to quantify the contribution of workers in completing sensing tasks. Meanwhile, to ensure sensing data quality and worker selection fairness, we propose a Fairness-based Truth Discovery Mechanism (FTDM) with two core modules: a fine-grained reputation system to identify reliable workers and filter out malicious ones, and an upper confidence bound algorithm to optimize worker selection and avoid local optima. Finally, we implement these functions in BSFOIMToken and deploy a prototype on the Ethereum blockchain, demonstrating its practicality and robust performance. Rigorous theoretical and comprehensive experimental tests have proven their adherence to truthfulness, budget feasibility and individual rationality.

Federated learning, hailed as a privacy-preserving collaboration paradigm, has garnered significant attention in research circles. Typically, it involves multiple clients collaborating to integrate multi-party knowledge, facilitating the learning of a shared global model with decentralized local data. Despite the popularity of federated learning, the surge in approaches addressing various realistic challenges has highlighted a critical issue. The aggregated model may struggle to capture diverse domain knowledge across participants, leading to limited performance in cross-client domain scenarios. Furthermore, the incorporation of knowledge from participating parties can hinder generalization on out-of-client distributions. To comprehensively address this challenge, we dissect federated generalization into two dimensions: the participating domain and the unseen domain. In this paper, we propose a novel solution incorporating domain-specialized and invariant experts. These experts are designed to faithfully represent individual domain characteristics and different domain universality. Additionally, we introduce a pioneering test-time expert aggregation strategy that utilizes prediction consistency metrics to aggregate different experts, specifically tailored for handling agnostic testing distributions. Empirical results validate that our proposed methodology significantly enhances federated performance on both cross-client and out-of-client generalization under different scenarios and with various related methods. A comprehensive ablation study demonstrates the effectiveness of the proposed modules.

Text-based Person Retrieval (TPR) plays a pivotal role in video surveillance systems for safeguarding public safety. As a fine-grained retrieval task, TPR faces the significant challenge of precisely capturing highly discriminative features across image and text modalities. Existing methods primarily focus on establishing modality-shared feature spaces to bridge cross-modal discrepancies. However, these methods are prone to disturbances from irrelevant information, such as background noises in the visual modality, and often over-emphasize specific local regions while neglecting the capture of diverse discriminative modal features, thereby limiting the robustness of cross-modal matching. In this paper, we introduce a novel framework, termed the Diverse Co-saliency Feature Learning Network (DCFL), which mines the co-saliency information between image and text modalities and enhances the diversity of cross-modal discriminative features while mitigating the interference of noise. Specifically, to construct cross-modal co-saliency features, we devise the Intra-modal Saliency Feature Learning (ISFL) and Cross-modal Saliency Feature Matching (CSFM) modules. ISFL employs a weighted mask mechanism to guide the model in reducing the impact of noise information in both modalities. Complementing ISFL, CSFM establishes consistent relationships between saliency features across modalities, leveraging text descriptions to align pedestrian-relevant visual regions. Furthermore, we propose the Diverse Co-saliency Feature Mining (DCFM) to bolster the diversity of discriminative co-saliency features across both image and text modalities. This module integrates a diversity regularization term, enabling the extraction of varied visual cues and capturing comprehensive features of the target individual. Extensive benchmark experiments demonstrate a substantial superiority of our approach over the state-of-the-art methods. The code will be released publicly.

With the rapid development and widespread adoption of digital media, deceptive behaviors have raised numerous ethical and security issues, making the research and advancement of deception detection technology particularly important. Most previous automated deception detection methods primarily focus on facial information in a visual context. However, from a psychological perspective, deceptive behavior extends beyond mere changes in facial expressions; it can also manifest through limb behaviors and subtle incoordination among body components. Motivated by this inconsistency, this paper attempts to model body behaviors and their relationships for deception detection. It is worth noting that some mainstream video understanding methods can roughly model head and limb information, but their holistic video input approach is easily affected by background interference. This limits their ability to focus on key body regions and subtle motion cues that reflect deception, thereby restricting detection performance. To address the above challenges, this paper proposes a Dynamic Learning Framework leveraging Body Part Relationship-Aware Modeling (DLF-BRAM). Within this framework, we segment and model the head and limb regions to reduce irrelevant background interference and enhance the accuracy of feature learning. The framework includes two main components: the Head-Limb Relationship-Aware Representation (HLRAR) module and the Dynamic Assessment Learning Strategy (DALS). The HLRAR module reveals the spatiotemporal relationship of the head, limbs, and their interactions, and learns deep feature representations for each cue, thereby highlighting the uniqueness of these cues. DALS evaluates the learning effectiveness of the three spatiotemporal relationships during training and dynamically adjusts their learning weights, preventing dominance by any single branch and promoting balanced learning. Extensive benchmark and ablation experiments demonstrate that our method outperforms most existing approaches, verifying its effectiveness.

Unsupervised object re-identification (Re-ID) aims to learn discriminative features without identity annotations. Existing mainstream methods are usually developed based on convolutional neural networks for feature extraction and pseudo-label estimation. However, convolutional neural networks suffer from limitations in capturing dispersed long-range dependencies and integrating global information. In comparison, vision transformers demonstrate superior robustness in complex environments, leveraging their versatile modeling capabilities to process diverse data structures with greater precision. In this paper, we delve into the potential of vision transformers in unsupervised Re-ID, proposing a Transformer-based perception-assisted framework (PAT). Considering Re-ID is a typical fine-grained task, existing unsupervised Re-ID methods relying on pseudo-labels generated by clustering algorithms provide only category-level discriminative supervision, with limited attention to local details. Therefore, we propose a novel target-aware mask alignment (TMA) strategy that provides additional supervision signals by leveraging low-level visual cues. Specifically, we employ pseudo-labels to guide the fine-grained alignment of features with local pixel information from critical discriminative regions. This method establishes a mutual learning mechanism via a shared Transformer, effectively balancing discriminative learning and detailed understanding. Furthermore, we propose a perceptual fusion feature augmentation (PFA) method to optimize instance-level discriminative learning. The proposed method is evaluated on multiple Re-ID datasets, demonstrating superior performance and robustness in comparison to state-of-the-art techniques. Notably, without annotations, our method achieves better results than many supervised counterparts. The code will be released.

Text-based person search (TBPS) aims to retrieve images of a specific person from a large image gallery based on a natural language description. Existing methods rely on massive annotated image-text data to achieve satisfactory performance in fully-supervised learning. This presents a substantial practical challenge, given the difficulty in obtaining annotated texts for person images. This work undertakes a pioneering initiative to explore TBPS under the semi-supervised setting, where only a limited number of person images are annotated with textual descriptions while the majority of images lack annotations. We present a two-stage basic solution based on generation-then-retrieval for semi-supervised TBPS. The generation stage enriches annotated data by applying an image captioning model to generate pseudo-texts for unannotated images. Later, the retrieval stage performs fully-supervised retrieval learning using the augmented data. Crucially, considering the noise interference of the pseudo-texts on retrieval learning, we propose a noise-robust retrieval framework that enhances the ability of the retrieval model to handle noisy data. The framework integrates two key strategies: Hybrid Patch-Channel Masking (PC-Mask) to refine the model architecture, and Noise-Guided Progressive Training (NP-Train) to enhance the training process. PC-Mask performs masking on the input data at both the patch-level and the channel-level to prevent overfitting noisy supervision. NP-Train introduces a progressive training schedule based on the noise level of pseudo-texts to facilitate noise-robust learning. Extensive experiments on multiple TBPS benchmarks show that the proposed framework achieves promising performance under the semi-supervised setting.

Multimodal emotion recognition is a task that integrates textual, visual, and audio data to holistically infer an individual's emotional state. Existing research predominantly focuses on exploiting modality-specific cues for joint learning, often ignoring the differences between multiple modalities in common goal learning. Due to multimodal heterogeneity, common goal learning inadvertently introduces optimization biases and interaction noise. To address above challenges, we propose a novel approach named Gradient and Structure Consistency (GSCon). Our strategy operates at both overall and individual levels to consider balance optimization and effective interaction respectively. At the overall level, to avoid the optimization suppression of one modality on others, we construct a balanced gradient direction that aligns each modality's optimization direction, ensuring unbiased convergence. Simultaneously, at the individual level, to avoid the interaction noise caused by multimodal alignment, we align the spatial structure of samples in different modalities. The spatial structure of the samples will not differ due to modal heterogeneity, achieving effective inter-modal interaction. Extensive experiments on multimodal emotion recognition and multimodal intention understanding datasets demonstrate the effectiveness of the proposed method. Code is available at https://github.com/ShiQingHongYa/GSCon.

Segmentation of cell nuclei from three-dimensional (3D) volumetric fluorescence microscopy images is crucial for biological and clinical analyses. In recent years, convolutional neural networks have become the reliable 3D medical image segmentation standard. However, convolutional layers are limited by their finite receptive fields and weight-sharing mechanisms. Consequently, they struggle to effectively model long-range dependencies and spatial correlations, which may lead to inadequate nuclei segmentation. Moreover, the diversity in nuclear appearance and density poses additional challenges. This work proposes a lightweight multi-layer deep aggregation network, MLDA-Net, incorporating Wide Receptive Field Attention (WRFA). This module effectively simulates the large receptive field generated by self-attention in the Swin Transformer while requiring fewer model parameters. This design implements an extended global sensory field that enhances the ability to capture a wide range of spatial information. In addition, the multiple cross-attention (MCA) module in MLDA-Net enhances the output features of different resolutions from the encoder while maintaining global effectiveness. The Multi-Path Aggregation Feature Pyramid Network (MAFPN) receives multi-scale outputs from the MCA module, generating a robust hierarchical feature pyramid for the final prediction. MLDA-Net outperforms state-of-the-art networks, including 3DU-Net, nnFormer, UNETR, SwinUNETR, and 3DUXNET, on the 3D volumetric datasets NucMM and MitoEM. It achieves average performance improvements of 4% to 7% in F1 score, MIoU, and PQ metrics, thereby establishing new benchmark results.

Federated Learning (FL) enables multiple devices to collaboratively train models without sharing their raw data. Considering that clients may prefer to design their own models independently, model heterogeneous FL has emerged. Additionally, due to the annotation uncertainty, the collected data usually contain unavoidable and varying noise, which cannot be effectively addressed by existing FL algorithms. This paper presents a novel solution that simultaneously handles model heterogeneity and label noise in a single framework. It is featured in three aspects: (1) For the communication between heterogeneous models, we directly align the model feedback by utilizing the easily-accessible public data, which does not require additional global models or relevant data for collaboration. (2) For internal label noise in each client, we design a dynamic label refinement strategy to mitigate the negative effects. (3) For challenging noisy feedback from other participants, we design an enhanced client confidence re-weighting scheme, which adaptively assigns corresponding weights to each client in the collaborative learning stage. Extensive experiments validate the effectiveness of our approach in mitigating the negative effects of various noise rates and types under both model homogeneous and heterogeneous FL settings.

Vertical Federated Learning (VFL) has emerged as a crucial privacy-preserving learning paradigm that involves training models using distributed features from shared samples. However, the performance of VFL can be hindered when the number of shared or aligned samples is limited, a common issue in mobile environments where user data are diverse and unaligned across multiple devices. Existing approaches use feature generation and pseudo-label estimation for unaligned samples to address this issue, unavoidably introducing noise during the generation process. In this work, we propose Local Enhanced Effective Vertical Federated Learning (LEEF-VFL), which fully utilizes unaligned samples in the local learning before collaboration. Unlike previous methods that overlook private labels owned by each client, we leverage these private labels to learn from all local samples, constructing robust local models to serve as solid foundations for collaborative learning. Additionally, we reveal that the limited number of aligned samples introduces distribution bias from global data distribution. In this case, we propose to minimize the distribution discrepancies between the aligned samples and the global data distribution to enhance collaboration. Extensive experiments demonstrate the effectiveness of LEEF-VFL in addressing the challenges of limited aligned samples, making it suitable for VFL in mobile computing environments.

Existing deep learning methods have achieved remarkable results in diagnosing retinal diseases, showcasing the potential of advanced AI in ophthalmology. However, the black-box nature of these methods obscures the decision-making process, compromising their trustworthiness and acceptability. Inspired by the concept-based approaches and recognizing the intrinsic correlation between retinal lesions and diseases, we regard retinal lesions as concepts and propose an inherently interpretable framework designed to enhance both the performance and explainability of diagnostic models. Leveraging the transformer architecture, known for its proficiency in capturing long-range dependencies, our model can effectively identify lesion features. By integrating with image-level annotations, it achieves the alignment of lesion concepts with human cognition under the guidance of a retinal foundation model. Furthermore, to attain interpretability without losing lesion-specific information, our method employs a classifier built on a cross-attention mechanism for disease diagnosis and explanation, where explanations are grounded in the contributions of human-understandable lesion concepts and their visual localization. Notably, due to the structure and inherent interpretability of our model, clinicians can implement concept-level interventions to correct the diagnostic errors by simply adjusting erroneous lesion predictions. Experiments conducted on four fundus image datasets demonstrate that our method achieves favorable performance against state-of-the-art methods while providing faithful explanations and enabling concept-level interventions. Our code is publicly available at https://github.com/Sorades/CLAT.

This article studies an emerging practical problem called heterogeneous prototype learning (HPL). Unlike the conventional heterogeneous face synthesis (HFS) problem that focuses on precisely translating a face image from a source domain to another target one without removing facial variations, HPL aims at learning the variation-free prototype of an image in the target domain while preserving the identity characteristics. HPL is a compounded problem involving two cross-coupled subproblems, that is, domain transfer and prototype learning (PL), thus making most of the existing HFS methods that simply transfer the domain style of images unsuitable for HPL. To tackle HPL, we advocate disentangling the prototype and domain factors in their respective latent feature spaces and then replacing the source domain with the target one for generating a new heterogeneous prototype. In doing so, the two subproblems in HPL can be solved jointly in a unified manner. Based on this, we propose a disentangled HPL framework, dubbed DisHPL, which is composed of one encoder-decoder generator and two discriminators. The generator and discriminators play adversarial games such that the generator embeds contaminated images into a prototype feature space only capturing identity information and a domain-specific feature space, while generating realistic-looking heterogeneous prototypes. Experiments on various heterogeneous datasets with diverse variations validate the superiority of DisHPL.

Text-based Person Search (TBPS) aims to retrieve the person images using natural language descriptions. Recently, Contrastive Language Image Pretraining (CLIP), a universal large cross-modal vision-language pre-training model, has remarkably performed over various cross-modal downstream tasks due to its powerful cross-modal semantic learning capacity. TPBS, as a fine-grained cross-modal retrieval task, is also facing the rise of research on the CLIP-based TBPS. In order to explore the potential of the visual-language pre-training model for downstream TBPS tasks, this paper makes the first attempt to conduct a comprehensive empirical study of CLIP for TBPS and thus contribute a straightforward, incremental, yet strong TBPS-CLIP baseline to the TBPS community. We revisit critical design considerations under CLIP, including data augmentation and loss function. The model, with the aforementioned designs and practical training tricks, can attain satisfactory performance without any sophisticated modules. Also, we conduct the probing experiments of TBPS-CLIP in model generalization and model compression, demonstrating the effectiveness of TBPS-CLIP from various aspects. This work is expected to provide empirical insights and highlight future CLIP-based TBPS research.

Federated Graph Learning is a privacy-preserving collaborative approach for training a shared model on graph-structured data in the distributed environment. However, in real-world scenarios, the client graph data usually originate from diverse domains, this unavoidably hinders the generalization performance of the final global model. To address this challenge, we start the first attempt to investigate this scenario by learning a well-generalizable model. In order to improve the performance of the global model from different perspectives, we propose a novel framework called Federated Graph Learning with Generalizable Prototypes (FGGP). It decouples the global model into two levels and bridges them via prototypes. These prototypes, which are semantic centers derived from the feature extractor, can provide valuable classification information. At the classification model level, we innovatively eschew the traditional classifiers, then instead leverage clustered prototypes to capture fruitful domain information and enhance the discriminative capability of the classes, improving the performance of multi-domain predictions. Furthermore, at the feature extractor level, we go beyond traditional approaches by implicitly injecting distinct global knowledge and employing contrastive learning to obtain more powerful prototypes while enhancing the feature extractor generalization ability. Experimental results on various datasets are presented to validate the effectiveness of the proposed method.

Fashion image editing aims to edit an input image to obtain richer or distinct visual clothing matching effects. Existing global fashion image editing methods are difficult to achieve rich outfit combination effects while local fashion image editing is more in line with the needs of diverse and personalized outfit matching. The local editing techniques typically depend on text and auxiliary modalities (e.g., human poses, human keypoints, garment sketches, etc.) for image manipulation, where the auxiliary modalities essentially assist in locating the editing region. Since these auxiliary modalities usually involve additional efforts in practical application scenarios, text-driven fashion image editing shows high flexibility. In this paper, we propose TexFit, a Text-driven Fashion image Editing method using diffusion models, which performs the local image editing only with the easily accessible text. Our approach employs a text-based editing region location module to predict precise editing region in the fashion image. Then, we take the predicted region as the generation condition of diffusion models together with the text prompt to achieve precise local editing of fashion images while keeping the rest part intact. In addition, previous fashion datasets usually focus on global description, lacking local descriptive information that can guide the precise local editing. Therefore, we develop a new DFMM-Spotlight dataset by using region extraction and attribute combination strategies. It focuses locally on clothes and accessories, enabling local editing with text input. Experimental results on the DFMM-Spotlight dataset demonstrate the effectiveness of our model. Code and Datasets are available at https://texfit.github.io/.

Cross-modal noise-robust learning is a challenging task since noisy correspondence is hard to recognize and rectify. Due to the cumulative and unavoidable negative impact of unresolved noise, existing methods cannot maintain a stable performance when the noise increases. In this paper, we present a novel Negative Pre-aware Cross-modal (NPC) matching solution for large visual-language model fine-tuning on noisy downstream tasks. It is featured in two aspects: (1) For noise recognition and resistance, previous methods usually directly filter out a noise subset, we propose to estimate the negative impact of each sample. It does not need additional correction mechanisms that may predict unreliable correction results, leading to self-reinforcing error. We assign a confidence weight to each sample according to its negative impact in the training process. This adaptively adjusts the contribution of each sample to avoid noisy accumulation. (2) For maintaining stable performance with increasing noise, we utilize the memorization effect of DNNs by maintaining a memory bank. Specifically, we apply GMM to select high-confident clean samples as the memory entry, where the memory entry is used to estimate the negative impact of each sample. Since clean samples are easier distinguished by GMM with increasing noise, the memory bank can still maintain high quality at a high noise ratio. Compared to the correction mechanism focusing on noise samples, memory bank-based estimation is more robust, which makes the model performance stable on noisy datasets. Extensive experiments demonstrate that our method significantly improves matching accuracy and performance stability at increasing noise ratio. Our approach also surpasses the state-of-the-art methods by a large margin. The code is available at: https://github.com/ZhangXu0963/NPC.

Federated learning (FL) has emerged as a new paradigm for privacy-preserving collaborative training. Under domain skew, the current FL approaches are biased and face two fairness problems. 1) Parameter Update Conflict: data disparity among clients leads to varying parameter importance and inconsistent update directions. These two disparities cause important parameters to potentially be over-whelmed by unimportant ones of dominant updates. It consequently results in significant performance decreases for lower-performing clients. 2) Model Aggregation Bias: existing FL approaches introduce unfair weight allocation and neglect domain diversity. It leads to biased model convergence objective and distinct performance among domains. We discover a pronounced directional update consistency in Federated Learning and propose a novel framework to tackle above issues. First, leveraging the discovered characteristic, we selectively discard unimportant parameter updates to prevent updates from clients with lower performance overwhelmed by unimportant parameters, resulting in fairer generalization performance. Second, we propose a fair aggregation objective to prevent global model bias towards some domains, ensuring that the global model continuously aligns with an unbiased model. The proposed method is generic and can be combined with other existing FL methods to enhance fairness. Comprehensive experiments on Digits and Office-Caltech demonstrate the high fairness and performance of our method.

In Re-identification (ReID), recent advancements yield noteworthy progress in both unimodal and cross-modal re-trieval tasks. However, the challenge persists in developing a unified framework that could effectively handle varying multimodal data, including RGB, infrared, sketches, and textual information. Additionally, the emergence of large-scale models shows promising performance in various vision tasks but the foundation model in ReID is still blank. In response to these challenges, a novel multimodal learning paradigm for ReID is introduced, referred to as All-in-One (AIO), which harnesses a frozen pre-trained big model as an encoder, enabling effective multimodal re-trieval without additional fine-tuning. The diverse multi-modal data in AIO are seamlessly tokenized into a unified space, allowing the modality-shared frozen encoder to extract identity-consistent features comprehensively across all modalities. Furthermore, a meticulously crafted ensemble of cross-modality heads is designed to guide the learning trajectory. AIO is the first framework to perform all-in-one ReID, encompassing four commonly used modali-ties. Experiments on cross-modal and multimodal ReID reveal that AIO not only adeptly handles various modal data but also excels in challenging contexts, showcasing exceptional performance in zero-shot and domain generalization scenarios. Code will be available at: https://github.com/lihe404/AIO.

Multimodal intent recognition (MIR) aims to perceive the human intent polarity via language, visual, and acoustic modalities. The inherent intent ambiguity makes it challenging to recognize in multimodal scenarios. Existing MIR methods tend to model the individual video independently, ignoring global contextual information across videos. This learning manner inevitably introduces perception biases, exacerbated by the inconsistencies of the multimodal representation, amplifying the intent uncertainty. This challenge motivates us to explore effective global context modeling. Thus, we propose a context-augmented global contrast (CAGC) method to capture rich global context features by mining both intra-and cross-video context interactions for MIR. Concretely, we design a context-augmented transformer module to extract global context dependencies across videos. To further alleviate error accumulation and interference, we develop a cross-video bank that retrieves effective video sources by considering both intentional tendency and video similarity. Furthermore, we introduce a global context-guided contrastive learning scheme, designed to mitigate inconsistencies arising from global context and individual modalities in different feature spaces. This scheme incorporates global cues as the supervision to capture robust the multimodal intent representation. Experiments demonstrate CAGC obtains superior performance than state-of-the-art MIR methods. We also generalize our approach to a closely related task, multimodal sentiment analysis, achieving the comparable performance.

Unsupervised visible-infrared person re-identification (US-VI-ReID) centers on learning a cross-modality retrieval model without labels, reducing the reliance on expensive cross-modality manual annotation. Previous US-VI-ReID works gravitate toward learning cross-modality information with the deep features extracted from the ultimate layer. Nevertheless, interfered by the multiple discrepancies, solely relying on deep features is insufficient for accurately learning modality-invariant features, resulting in negative optimization. The shallow feature from the shallow layers contains nuanced detail information, which is critical for effective cross-modality learning but is dis- regarded regrettably by the existing methods. To address the above issues, we design a Shallow-Deep Collaborative Learning (SDCL) framework based on the transformer with shallow-deep contrastive learning, incorporating Collaborative Neighbor Learning (CNL) and Collaborative Ranking Association (CRA) module. Specifically, CNL unveils the intrinsic homogeneous and heterogeneous collaboration which are harnessed for neighbor alignment, enhancing the robustness in a dynamic manner. Furthermore, CRA associates the cross-modality labels with the ranking association between shallow and deep features, furnishing valuable supervision for cross-modality learning. Extensive experiments validate the superiority of our method, even outperforming certain supervised counterparts.

Personalized Federated Learning (PFL) is primarily de-signed to provide customized models for each client to better fit the non-iid distributed client data, which is a inherent challenge in Federated Learning. However, current PFL methods suffer from inconsistencies in both intra-client and inter-client levels: 1) The intra-client inconsistency stems from the asynchronous update strategy for personalized and shared parameters. In PFL, clients update their shared parameters to communicate and learn from others, while keeping personalized parts unchanged, leading to poor coordination between these two components. 2) The Inter-client inconsistency arises from “stragglers” - inactive clients that communicate and train with the server less frequently. This results in their under-trained personalized models and impedes the collaborative training stage for other clients. In this paper, we present a novel PFL framework named FedAS, which uses Federated Parameter-Alignment and Client-Synchronization to overcome above challenges. Initially, we enhance the localization of global parameters by infusing them with local insights. We make the shared parts learn from previous model, thereby increasing their local relevance and reducing the impact of parameter inconsistency. Furthermore, we design a robust aggregation method to mitigate the impact of stragglers by preventing the incorporation of their under-trained knowledge into aggregated model. Experimental results on CifarlO and CifarlOO validate the effectiveness of our FedAS in achieving better performance and robustness against data heterogeneity.

Federated learning presents massive potential for privacy-friendly collaboration. However, the performance of federated learning is deeply affected by byzantine attacks, where malicious clients deliberately upload crafted vicious updates. While various robust aggregations have been proposed to defend against such attacks, they are subject to certain assumptions: homogeneous private data and related proxy datasets. To address these limitations, we propose Self-Driven Entropy Aggregation (SDEA), which leverages the random public dataset to conduct Byzantine-robust aggregation in heterogeneous federated learning. For Byzantine attackers, we observe that benign ones typically present more confident (sharper) predictions than evils on the public dataset. Thus, we highlight benign clients by introducing learnable aggregation weight to minimize the instance-prediction entropy of the global model on the random public dataset. Besides, with inherent data heterogeneity in federated learning, we reveal that it brings heterogeneous sharpness. Specifically, clients are optimized under distinct distribution and thus present fruitful predictive preferences. The learnable aggregation weight blindly allocates high attention to limited ones for sharper predictions, resulting in a biased global model. To alleviate this problem, we encourage the global model to offer diverse predictions via batch-prediction entropy maximization and conduct clustering to equally divide honest weights to accommodate different tendencies. This endows SDEA to detect Byzantine attackers in heterogeneous federated learning. Empirical results demonstrate the effectiveness.

Graph Contrastive Learning (GCL) has emerged as a highly effective self-supervised approach in graph representation learning. However, prevailing GCL methods confront two primary challenges: 1) They predominantly operate under homophily assumptions, focusing on low-frequency signals in node features while neglecting heterophilic edges that connect nodes with dissimilar features. 2) Their reliance on neighborhood aggregation for inference leads to scalability challenges and hinders deployment in real-time applications. In this paper, we introduce S3GCL, an innovative framework designed to tackle these challenges. Inspired by spectral GNNs, we initially demonstrate the correlation between frequency and homophily levels. Then, we propose a novel cosine-parameterized Chebyshev polynomial as low/high-pass filters to generate biased graph views. To resolve the inference dilemma, we incorporate an MLP encoder and enhance its awareness of graph context by introducing structurally and semantically neighboring nodes as positive pairs in the spatial domain. Finally, we formulate a cross-pass GCL objective between full-pass MLP and biased-pass GNN filtered features, eliminating the need for augmentation. Extensive experiments on real-world tasks validate S3GCL proficiency in generalization to diverse homophily levels and its superior inference efficiency.

Person re-identification (ReID) is crucial in video surveillance, aiming to match individuals across different camera views while cloth-changing person re-identification (CC-ReID) focuses on pedestrians changing attire. Many existing CC-ReID methods overlook generalization, crucial for universality across cloth-consistent and cloth-changing scenarios. This paper pioneers exploring the cloth-generalized person re-identification (CG-ReID) task and introduces the Cloth-aware Augmentation (CaAug) strategy. Comprising domain augmentation and feature augmentation, CaAug aims to learn identity-relevant features adaptable to both scenarios. Domain augmentation involves creating diverse fictitious domains and simulating various clothing scenarios. Supervising features from different cloth domains enhances robustness and generalization against clothing changes. Additionally, for feature augmentation, element exchange introduces diversity concerning clothing changes. Regularizing the model with these augmented features strengthens resilience against clothing change uncertainty. Extensive experiments on cloth-changing datasets demonstrate the efficacy of our approach, consistently outperforming state-of-the-art methods.

Graph Neural Networks (GNNs) are widely employed to derive meaningful node representations from graphs. Despite their success, deep GNNs frequently grapple with the oversmoothing issue, where node representations become highly indistinguishable due to repeated aggregations. In this work, we consider the oversmoothing issue from two aspects of the node embedding space: dimension and instance. Specifically, while existing methods primarily concentrate on instance-level node relations to mitigate oversmoothing, we propose to mitigate oversmoothing at dimension level. We reveal the heightened information redundancy between dimensions which diminishes information diversity and impairs node differentiation in GNNs. Motivated by this insight, we propose the Dimension-Level Decoupling (DLD) to reduce dimension redundancy, enhancing dimensional-level node differentiation. Besides, at the instance level, the neglect of class differences leads to vague classification boundaries. Hence, we introduce the Instance-Level Class-Difference Decoupling (ICDD) that repels inter-class nodes and attracts intra-class nodes, improving the instance-level node discrimination with clear classification boundaries. Additionally, we introduce a novel evaluation metric that considers the impact of class differences on node distances, facilitating precise oversmoothing measurement. Extensive experiments demonstrate the effectiveness of our method Dual-Dimensional Class-Difference Decoupling (DDCD) across diverse scenarios.

Visible-Infrared Person Re-identification (VI-ReID) is a challenging cross-modal retrieval task due to significant modality differences, primarily resulting from the absence of color information in the infrared modality. The development of large foundation models like Large Language Models (LLMs) and Vision Language Models (VLMs) motivates us to explore a feasible solution to empower VI-ReID with off-the-shelf large foundation models. To this end, we propose a novel Text-enhanced VI-ReID framework driven by Large Foundation Models (TVI-LFM). The core idea is to enrich the representation of the infrared modality with textual descriptions automatically generated by VLMs. Specifically, we incorporate a pre-trained VLM to extract textual features from texts generated by VLM and augmented by LLM, and incrementally fine-tune the text encoder to minimize the domain gap between generated texts and original visual modalities. Meanwhile, to enhance the infrared modality with extracted textual representations, we leverage modality alignment capabilities of VLMs and VLM-generated feature-level filters. This enables the text model to learn complementary features from the infrared modality, ensuring the semantic structural consistency between the fusion modality and the visible modality. Furthermore, we introduce modality joint learning to align features across all modalities, ensuring that textual features maintain stable semantic representation of overall pedestrian appearance during complementary information learning. Additionally, a modality ensemble retrieval strategy is proposed to leverage complementary strengths of each query modality to improve retrieval effectiveness and robustness. Extensive experiments on three expanded VI-ReID datasets demonstrate that our method significantly improves the retrieval performance, paving the way for the utilization of large foundation models in downstream multi-modal retrieval tasks.

Backdoor attacks pose a serious threat to federated systems, where malicious clients optimize on the triggered distribution to mislead the global model towards a predefined target. Existing backdoor defense methods typically require either homogeneous assumption, validation datasets, or client optimization conflicts. In our work, we observe that benign heterogeneous distributions and malicious triggered distributions exhibit distinct parameter importance degrees. We introduce the Fisher Discrepancy Cluster and Rescale (FDCR) method, which utilizes Fisher Information to calculate the degree of parameter importance for local distributions. This allows us to reweight client parameter updates and identify those with large discrepancies as backdoor attackers. Furthermore, we prioritize rescaling important parameters to expedite adaptation to the target distribution, encouraging significant elements to contribute more while diminishing the influence of trivial ones. This approach enables FDCR to handle backdoor attacks in heterogeneous federated learning environments. Empirical results on various heterogeneous federated scenarios under backdoor attacks demonstrate the effectiveness of our method.

Non-exemplar class-incremental learning (NECIL) is a challenging task that requires recognizing both old and new classes without retaining any old class samples. Current works mainly deal with the conflicts between old and new classes retrospectively as a new task comes in. However, the lack of old task data makes balancing old and new classes difficult. Instead, we propose a Prospective Representation Learning (PRL) approach to prepare the model for handling conflicts in advance. In the base phase, we squeeze the embedding distribution of the current classes to reserve space for forward compatibility with future classes. In the incremental phase, we make the new class features away from the saved prototypes of old classes in a latent space while aligning the current embedding space with the latent space when updating the model. Thereby, the new class features are clustered in the reserved space to minimize the shock of the new classes on the former classes. Our approach can help existing NECIL baselines to balance old and new classes in a plug-and-play manner. Extensive experiments on several benchmarks demonstrate that our approach outperforms the state-of-the-art methods.

Personalized Federated Graph Learning (pFGL) facilitates the decentralized training of Graph Neural Networks (GNNs) without compromising privacy while accommodating personalized requirements for non-IID participants. In cross-domain scenarios, structural heterogeneity poses significant challenges for pFGL. Nevertheless, previous pFGL methods incorrectly share non-generic knowledge globally and fail to tailor personalized solutions locally under domain structural shift. We innovatively reveal that the spectral nature of graphs can well reflect inherent domain structural shifts. Correspondingly, our method overcomes it by sharing generic spectral knowledge. Moreover, we indicate the biased message-passing schemes for graph structures and propose the personalized preference module. Combining both strategies, we propose our pFGL framework $\textbf{FedSSP}$ which $\textbf{S}$hares generic $\textbf{S}$pectral knowledge while satisfying graph $\textbf{P}$references. Furthermore, We perform extensive experiments on cross-dataset and cross-domain settings to demonstrate the superiority of our framework. The code is available at https://github.com/OakleyTan/FedSSP.

Cross-modality visible-infrared person reidentification (VI-ReID) has attracted widespread concern due to its scalability in 24-h video surveillance of the Visual Internet of Things (VIoT). Driven by enough annotated training data, supervised VI-ReID has achieved superior performance. However, annotating a large amount of cross-modality data is extremely time-consuming, which limits its employment in real-world scenarios. Existing several works neglect the image-level discrepancy and could not obtain reliable feature-level heterogeneous correlation. In this article, we propose a novel homogeneous and heterogeneous optimization with modality style adaptation (HHO) mechanism to eliminate intramodality and intermodality discrepancies without any label information for unsupervised VI-ReID. Specifically, we present the modality style adaptation strategy to transfer unlabeled cross-modality pedestrian styles, which not only increases the image diversity but also bridges the intermodality gap. Meanwhile, we employ the clustering algorithm to generate pseudo labels for each modality. The homogeneous feature optimization is developed to extract intramodality pedestrian features. Furthermore, we propose heterogeneous feature optimization to eliminate the intermodality discrepancy. To this end, a heterogeneous feature search (HFS) module is designed to mine reliable cross-modality signals for each identity. These reliable heterogeneous features are constrained to generate the compact feature distribution, while different identities are forced to be separated. The HHO are seamlessly integrated to learn cross-modality robust features. Abundant experiments prove the superiority of HHO, which gains superior performance.

Matching hand-drawn sketches with photos (a.k.a sketch-photo recognition or re-identification) faces the information asymmetry challenge due to the abstract nature of the sketch modality. Existing works tend to learn shared embedding spaces with CNN models by discarding the appearance cues for photo images or introducing GAN for sketch-photo synthesis. The former unavoidably loses discriminability, while the latter contains ineffaceable generation noise. In this paper, we start the first attempt to design an information-aligned sketch transformer (Sketch Trans+) viacross-modal disentangled prototype learning, while the transformer has shown great promise for discriminative visual modelling. Specifically, we design an asymmetric disentanglement scheme with a dynamic updatable auxiliary sketch (A-sketch) to align the modality representations without sacrificing information. The asymmetric disentanglement decomposes the photo representations into sketch-relevant and sketch-irrelevant cues, transferring sketch-irrelevant knowledge into the sketch modality to compensate for the missing information. Moreover, considering the feature discrepancy between the two modalities, we present a modality-aware prototype contrastive learning method that mines representative modality-sharing information using the modality-aware prototypes rather than the original feature representations. Extensive experiments on categoryand instance-level sketch-based datasets validate the superiority of our proposed method under various metrics. Code is available athttps://github.com/ccq195/SketchTrans

Federated learning is an important privacy-preserving multi-party learning paradigm, involving collaborative learning with others and local updating on private data. Model heterogeneity and catastrophic forgetting are two crucial challenges, which greatly limit the applicability and generalizability. This paper presents a novel FCCL+, federated correlation and similarity learning with non-target distillation, facilitating the both intra-domain discriminability and inter-domain generalization. For heterogeneity issue, we leverage irrelevant unlabeled public data for communication between the heterogeneous participants. We construct cross-correlation matrix and align instance similarity distribution on both logits and feature levels, which effectively overcomes the communication barrier and improves the generalizable ability. For catastrophic forgetting in local updating stage, FCCL+ introduces Federated Non Target Distillation, which retains inter-domain knowledge while avoiding the optimization conflict issue, fulling distilling privileged inter-domain information through depicting posterior classes relation. Considering that there is no standard benchmark for evaluating existing heterogeneous federated learning under the same setting, we present a comprehensive benchmark with extensive representative methods under four domain shift scenarios, supporting both heterogeneous and homogeneous federated settings. Empirical results demonstrate the superiority of our method and the efficiency of modules on various scenarios. The benchmark code for reproducing our results is available at https://github.com/WenkeHuang/FCCL.

Federated learning has emerged as a promising paradigm for privacy-preserving collaboration among different parties. Recently, with the popularity of federated learning, an influx of approaches have delivered towards different realistic challenges. In this survey, we provide a systematic overview of the important and recent developments of research on federated learning. First, we introduce the study history and terminology definition of this area. Then, we comprehensively review three basic lines of research: generalization, robustness, and fairness, by introducing their respective background concepts, task settings, and main challenges. We also offer a detailed overview of representative literature on both methods and datasets. We further benchmark the reviewed methods on several well-known datasets. Finally, we point out several open issues in this field and suggest opportunities for further research.

This paper introduces a simple yet powerful channel augmentation for visible-infrared re-identification. Most existing augmentation operations designed for single-modality visible images do not fully consider the imagery properties in visible to infrared matching. Our basic idea is to homogeneously generate color-irrelevant images by randomly exchanging the color channels. It can be seamlessly integrated into existing augmentation operations, consistently improving the robustness against color variations. For cross-modality metric learning, we design an enhanced channel-mixed learning strategy to simultaneously handle the intra- and cross-modality variations with squared difference for stronger discriminability. Besides, a weak-and-strong augmentation joint learning strategy is further developed to explicitly optimize the outputs of augmented images, which mutually integrates the channel augmented images (strong) and the general augmentation operations (weak) with consistency regularization. Furthermore, by conducting the label association between the channel augmented images and infrared modalities with modality-specific clustering, a simple yet effective unsupervised learning baseline is designed, which significantly outperforms existing unsupervised single-modality solutions. Extensive experiments with insightful analysis on two visible-infrared recognition tasks show that the proposed strategies consistently improve the accuracy. Without auxiliary information, the Rank-1/mAP achieves 71.48%/68.15% on the large-scale SYSU-MM01 dataset.

Prevalent federated learning commonly develops under the assumption that the ideal global class distributions are balanced. In contrast, real-world data typically follows the long-tailed class distribution, where models struggle to classify samples from tail classes. In this paper, we alleviate the issue under the long-tailed data, dissecting the into two aspects: the distorted feature space and the biased classifier. Specifically, we propose the Representation Unification and Classifier Rectification (RUCR), which leverages global unified prototypes to shape the feature space and calibrate the classifier. RUCR aggregates local prototypes (class-wise mean features) extracted by the global model to obtain global unified prototypes. It calibrates the feature space by pulling features within the same class towards corresponding global unified prototypes and pushing the other classes away. Moreover, RUCR utilizes global prototypes to reduce the classifier bias via prototypical mix-up. It generates a balanced virtual feature set by arbitrarily fusing global unified prototypes and local features. The classifier re-training is then conducted on the balanced virtual feature set to rectify the decision boundary and thus alleviate the shifts. Empirical results on CIFAR-10-LT, CIFAR-100-LT, and Tiny-Imagenet-LT datasets validate the superior performance of our proposed method.

Unsupervised visible-infrared person re-identification (US-VI-ReID) aims at learning a cross-modality matching model under unsupervised conditions, which is an extremely important task for practical nighttime surveillance to retrieve a specific identity. Previous advanced US-VI-ReID works mainly focus on associating the positive cross-modality identities to optimize the feature extractor by off-line manners, inevitably resulting in error accumulation of incorrect off-line cross-modality associations in each training epoch due to the intra-modality and inter-modality discrepancies. They ignore the direct cross-modality feature interaction in the training process, i.e., the on-line representation learning and updating. Worse still, existing interaction methods are also susceptible to inter-modality differences, leading to unreliable heterogeneous neighborhood learning. To address the above issues, we propose a dual consistency-constrained learning framework (DCCL) simultaneously incorporating off-line cross-modality label refinement and on-line feature interaction learning. The basic idea is that the relations between cross-modality instance-instance and instance-identity should be consistent. More specifically, DCCL constructs an instance memory, an identity memory, and a domain memory for each modality. At the beginning of each training epoch, DCCL explores the off-line consistency of cross-modality instance-instance and instance-identity similarities to refine the reliable cross-modality identities. During the training, DCCL finds credible homogeneous and heterogeneous neighborhoods with on-line consistency between query-instance similarity and query-instance domain probability similarities for feature interaction in one batch, enhancing the robustness against intra-modality and inter-modality variations. Extensive experiments validate that our method significantly outperforms existing works, and even surpasses some supervised counterparts. The source code is available athttps://github.com/yangbincv/DCCL.

Anonymization methods have gained widespread use in safeguarding privacy. However, conventional anonymization solutions inevitably lead to the loss of semantic information, resulting in limited data utility. Besides, existing deep learning-based anonymization strategies inadvertently alter the identities of pedestrians, rendering them unsuitable for re-identification (Re-ID) tasks. Beyond these limitations, we propose a joint learning reversible anonymization framework that can reversibly generate full-body anonymized images with little performance drop on Re-ID tasks. Despite these advancements, we reveal that the anonymization methods are vulnerable to model attacks, where attackers can utilize the anonymization model and public data to perform recovery and Re-ID tasks on anonymized images. To defend against the potential attack, we introduce the identity-specific encrypt-decrypt (ISED) architecture for enhanced security, where the anonymized images are encrypted using the specific key for each identity. It renders the images computationally inaccessible to attackers while allowing for seamless reversal without loss using the corresponding keys. Extensive experiments demonstrate that the anonymization framework can guarantee Re-ID performance while protecting pedestrian privacy. In addition, we provide both empirical and theoretical evidence to demonstrate the feasibility of model attacks and the effectiveness of our ISED strategy. Code is available at https://github.com/shentt67/SecureReID.

Visual intention understanding is a challenging task that explores the hidden intention behind the images of publishers in social media. Visual intention represents implicit semantics, whose ambiguous definition inevitably leads to label shifting and label blemish. The former indicates that the same image delivers intention discrepancies under different data augmentations, while the latter represents that the label of intention data is susceptible to errors or omissions during the annotation process. This paper proposes a novel method, called Label-aware Calibration and Relation-preserving (LabCR) to alleviate the above two problems from both intra-sample and inter-sample views. First, we disentangle the multiple intentions into a single intention for explicit distribution calibration in terms of the overall and the individual. Calibrating the class probability distributions in augmented instance pairs provides consistent inferred intention to address label shifting. Second, we utilize the intention similarity to establish correlations among samples, which offers additional supervision signals to form correlation alignments in instance pairs. This strategy alleviates the effect of label blemish. Extensive experiments have validated the superiority of the proposed method LabCR in visual intention understanding and pedestrian attribute recognition. Code is available at https://github.com/ShiQingHongYa/LabCR.

Automatic lesion segmentation is important for assisting doctors in the diagnostic process. Recent deep learning approaches heavily rely on large-scale datasets, which are difficult to obtain in many clinical applications. Leveraging external labelled datasets is an effective solution to tackle the problem of insufficient training data. In this paper, we propose a new framework, namely LatenTrans, to utilize existing datasets for boosting the performance of lesion segmentation in extremely low data regimes. LatenTrans translates non-target lesions into target-like lesions and expands the training dataset with target-like data for better performance. Images are first projected to the latent space via aligned style-based generative models, and rich lesion semantics are encoded using the latent codes. A novel consistency-aware latent code manipulation module is proposed to enable high-quality local style transfer from non-target lesions to target-like lesions while preserving other parts. Moreover, we propose a new metric, Normalized Latent Distance, to solve the question of how to select an adequate one from various existing datasets for knowledge transfer. Extensive experiments are conducted on segmenting lung and brain lesions, and the experimental results demonstrate that our proposed LatenTrans is superior to existing methods for cross-disease lesion segmentation.

The rising popularity of tabular data in data science applications has led to a surge of interest in utilizing deep neural networks (DNNs) to address tabular problems. Existing deep neural network methods are not effective in handling two fundamental challenges that are inherent in tabular data: permutation invariance (where the labels remain unchanged regardless of element order) and local dependency (where predictive labels are solely determined by local features). Furthermore, given the inherent heterogeneity among elements in tabular data, effectively capturing heterogeneous feature interactions remains unresolved. In this paper, we propose a novel Multiplex Cross-Feature Interaction Network (MPCFIN) by explicitly and systematically modeling feature relations with interactive graph neural networks. Specifically, MPCFIN first learns the most relevant features associated with individual features, and merges them to form cross-feature embedding. Subsequently, we design a multiplex graph neural network to learn enhanced representation for each sample. Comprehensive experiments on seven datasets demonstrate that MPCFIN exhibits superior performance over deep neural network methods in modeling the tabular data, showcasing consistent interpretability in its cross-feature embedding module for medical diagnosis applications.

This article presents a new adaptive metric distillation approach that can significantly improve the student networks' backbone features, along with better classification results. Previous knowledge distillation (KD) methods usually focus on transferring the knowledge across the classifier logits or feature structure, ignoring the excessive sample relations in the feature space. We demonstrated that such a design greatly limits performance, especially for the retrieval task. The proposed collaborative adaptive metric distillation (CAMD) has three main advantages: 1) the optimization focuses on optimizing the relationship between key pairs by introducing the hard mining strategy into the distillation framework; 2) it provides an adaptive metric distillation that can explicitly optimize the student feature embeddings by applying the relation in the teacher embeddings as supervision; and 3) it employs a collaborative scheme for effective knowledge aggregation. Extensive experiments demonstrated that our approach sets a new state-of-the-art in both the classification and retrieval tasks, outperforming other cutting-edge distillers under various settings.

Video captioning aims to generate natural language sentences that describe the given video accurately. Existing methods obtain favorable generation by exploring richer visual representations in encode phase or improving the decoding ability. However, the long-tailed problem hinders these attempts at low-frequency tokens, which rarely occur but carry critical semantics, playing a vital role in the detailed generation. In this paper, we introduce a novel Refined Semantic enhancement method towards Frequency Diffusion (RSFD), a captioning model that constantly perceives the linguistic representation of the infrequent tokens. Concretely, a Frequency-Aware Diffusion (FAD) module is proposed to comprehend the semantics of low-frequency tokens to break through generation limitations. In this way, the caption is refined by promoting the absorption of tokens with insufficient occurrence. Based on FAD, we design a Divergent Semantic Supervisor (DSS) module to compensate for the information loss of high-frequency tokens brought by the diffusion process, where the semantics of low-frequency tokens is further emphasized to alleviate the long-tailed problem. Extensive experiments indicate that RSFD outperforms the state-of-the-art methods on two benchmark datasets, i.e., MSR-VTT and MSVD, demonstrate that the enhancement of low-frequency tokens semantics can obtain a competitive generation effect. Code is available at https://github.com/lzp870/RSFD.

Person re-identification (ReID) with descriptive query (text or sketch) provides an important supplement for general image-image paradigms, which is usually studied in a single cross-modality matching manner, e.g., text-to-image or sketch-to-photo. However, without a camera-captured photo query, it is uncertain whether the text or sketch is available or not in practical scenarios. This motivates us to study a new and challenging modality-agnostic person re-ideruification problem. Towards this goal, we propose a unified person re-identification (UNIReID) architecture that can effectively adapt to cross-modality and multi-modality tasks. Specifically, UNIReID incorporates a simple dual-encoder with task-specific modality learning to mine and fuse visual and textual modality information. To deal with the imbalanced training problem of different tasks in UNIReID, we propose a task-aware dynamic training strategy in terms of task difficulty, adaptively adjusting the training focus. Besides, we construct three multi-modal ReID datasets by collecting the corresponding sketches from photos to support this challenging study. The experimental results on three multi-modal ReID datasets show that our UNIReID greatly improves the retrieval accuracy and generalization ability on different tasks and unseen scenarios.

Federated learning shows a bright promise as a privacy-preserving collaborative learning technique. However, prevalent solutions mainly focus on all private data sampled from the same domain. An important challenge is that when distributed data are derived from diverse domains. The private model presents degenerative performance on other domains (with domain shift). Therefore, we expect that the global model optimized after the federated learning process stably provides generalizability performance on multiple domains. In this paper, we propose Federated Proto-types Learning (FPL) for federated learning under domain shift. The core idea is to construct cluster prototypes and unbiased prototypes, providing fruitful domain knowledge and a fair convergent target. On the one hand, we pull the sample embedding closer to cluster prototypes belonging to the same semantics than cluster prototypes from distinct classes. On the other hand, we introduce consistency regularization to align the local instance with the respective unbiased prototype. Empirical results on Digits and Office Caltech tasks demonstrate the effectiveness of the proposed solution and the efficiency of crucial modules.

Text-to-image person retrieval aims to identify the target person based on a given textual description query. The primary challenge is to learn the mapping of visual and textual modalities into a common latent space. Prior works have attempted to address this challenge by leveraging separately pre-trained unimodal models to extract visual and textual features. However, these approaches lack the necessary underlying alignment capabilities required to match multimodal data effectively. Besides, these works use prior information to explore explicit part alignments, which may lead to the distortion of intra-modality information. To alleviate these issues, we present IRRA: a cross-modal Implicit Relation Reasoning and Aligning framework that learns relations between local visual-textual tokens and enhances global image-text matching without requiring additional prior supervision. Specifically, we first design an Implicit Relation Reasoning module in a masked language modeling paradigm. This achieves cross-modal interaction by integrating the visual cues into the textual tokens with a cross-modal multimodal interaction encoder. Secondly, to globally align the visual and textual embeddings, Similarity Distribution Matching is proposed to minimize the KL divergence between image-text similarity distributions and the normalized label matching distributions. The proposed method achieves new state-of-the-art results on all three public datasets, with a notable margin of about 3%-9% for Rank-1 accuracy compared to prior methods.

Unsupervised visible-infrared person re-identification is a challenging task due to the large modality gap and the unavailability of cross-modality correspondences. Cross-modality correspondences are very crucial to bridge the modality gap. Some existing works try to mine cross-modality correspondences, but they focus only on local information. They do not fully exploit the global relationship across identities, thus limiting the quality of the mined correspondences. Worse still, the number of clusters of the two modalities is often inconsistent, exacerbating the unreliability of the generated correspondences. In response, we devise a Progressive Graph Matching method to globally mine cross-modality correspondences under cluster imbalance scenarios. PGM formulates correspondence mining as a graph matching process and considers the global information by minimizing the global matching cost, where the matching cost measures the dissimilarity of clusters. Besides, PGM adopts a progressive strategy to address the imbalance issue with multiple dynamic matching processes. Based on PGM, we design an Alternate Cross Contrastive Learning (ACCL) module to reduce the modality gap with the mined cross-modality correspondences, while mitigating the effect of noise in correspondences through an alternate scheme. Extensive experiments demonstrate the reliability of the generated correspondences and the effectiveness of our method.

Visible modality and infrared modality person re-identification (VI-ReID) is an extremely important and challenging task. Existing works mainly focus on reducing the modality gap with Convolutional Neural Networks (CNN). However, the features extracted by CNN may contain useless identity-irrelevant information, which inevitably reduces the discrimination of features. To address this issue, this paper introduces a Top-K Visual Tokens Transformer (TVTR) framework which utilizes a top-k visual tokens selection module to accurately select top-k discriminative visual patches for reducing the distraction of identity-irrelevant information and learning discriminative features. Furthermore, a global-local circle loss is developed to optimize the TVTR for achieving cross-modality positive concentration and negative separation properties. The experimental results on SYSU-MM01 and RegDB datasets demonstrate the superiority of our method. The source code will be released.

Model heterogeneous federated learning is a realistic and challenging problem. However, due to the limitations of data collection, storage, and transmission conditions, as well as the existence of free-rider participants, the clients may suffer from data corruption. This paper starts the first attempt to investigate the problem of data corruption in the model heterogeneous federated learning framework. We design a novel method named Augmented Heterogeneous Federated Learning (AugHFL), which consists of two stages: 1) In the local update stage, a corruption-robust data augmentation strategy is adopted to minimize the adverse effects of local corruption while enabling the models to learn rich local knowledge. 2) In the collaborative update stage, we design a robust re-weighted communication approach, which implements communication between heterogeneous models while mitigating corrupted knowledge transfer from others. Extensive experiments demonstrate the effectiveness of our method in coping with various corruption patterns in the model heterogeneous federated learning setting.

Non-exemplar class-incremental learning (NECIL) requires deep models to maintain existing knowledge while continuously learning new classes without saving old class samples. In NECIL methods, prototypical representations are usually stored, which inject information from former classes to resist catastrophic forgetting in subsequent incremental learning. However, since the model continuously learns new knowledge, the stored prototypical representations cannot correctly model the properties of old classes in the existence of knowledge updates. To address this problem, we propose a novel prototype reminiscence mechanism that incorporates the previous class prototypes with arriving new class features to dynamically reshape old class feature distributions thus preserving the decision boundaries of previous tasks. In addition, to improve the model generalization on both newly arriving classes and old classes, we contribute an augmented asymmetric knowledge aggregation approach, which aggregates the overall knowledge of the current task and extracts the valuable knowledge of the past tasks, on top of self-supervised label augmentation. Experimental results on three benchmarks suggest the superior performance of our approach over the SOTA methods.

Unsupervised learning visible-infrared person re-identification (USL-VI-ReID) is an extremely important and challenging task, which can alleviate the issue of expensive cross-modality annotations. Existing works focus on handling the cross-modality discrepancy under unsupervised conditions. However, they ignore the fact that USL-VI-ReID is a cross-modality retrieval task with the hierarchical discrepancy, i.e., camera variation and modality discrepancy, resulting in clustering inconsistencies and ambiguous cross-modality label association. To address these issues, we propose a hierarchical framework to learn grand unified representation (GUR) for USL-VI-ReID. The grand unified representation lies in two aspects: 1) GUR adopts a bottom-up domain learning strategy with a cross-memory association embedding module to explore the information of hierarchical domains, i.e., intra-camera, inter-camera, and inter-modality domains, learning a unified and robust representation against hierarchical discrepancy. 2) To unify the identities of the two modalities, we develop a cross-modality label unification module that constructs a cross-modality affinity matrix as a bridge for propagating labels between two modalities. Then, we utilize the homogeneous structure matrix to smooth the propagated labels, ensuring that the label structure within one modality remains unchanged. Extensive experiments demonstrate that our GUR framework significantly outperforms existing USL-VI-ReID methods, and even surpasses some supervised counterparts.

Federated graph learning collaboratively learns a global graph neural network with distributed graphs, where the non-independent and identically distributed property is one of the major challenge. Most relative arts focus on traditional distributed tasks like images and voices, incapable of the graph structures. This paper firstly reveals that local client distortion is brought by both node-level semantics and graph-level structure. First, for node-level semantic, we find that contrasting nodes from distinct classes is beneficial to provide a well-performing discrimination. We pull the local node towards the global node of the same class and push them away from the global node of different classes. Second, we postulate that a well-structural graph neural network possesses similarity for neighbors due to the inherent adjacency relationships. However, aligning each node with adjacent nodes hinders discrimination due to the potential class inconsistency. We transform the adjacency relationships into the similarity distribution and leverage the global model to distill the relation knowledge into the local model, which preserves the structural information and discriminability of the local model. Empirical results on three graph datasets manifest the superiority of the proposed method over counterparts.

Personalized federated learning with differential privacy has been considered a feasible solution to address non-IID distribution of data and privacy leakage risks. However, current personalized federated learning methods suffer from inflexible personalization and convergence difficulties due to two main factors: 1) Firstly, we observe that the prevailing personalization methods mainly achieve this by personalizing a fixed portion of the model, which lacks flexibility. 2) Moreover, we further demonstrate that the default gradient calculation is sensitive to the widely-used clipping operations in differential privacy, resulting in difficulties in convergence. Considering that Fisher information values can serve as an effective measure for estimating the information content of parameters by reflecting the model sensitivity to parameters, we aim to leverage this property to address the aforementioned challenges. In this paper, we propose a novel federated learning method with Dynamic Fisher Personalization and Adaptive Constraint (FedDPA) to handle these challenges. Firstly, by using layer-wise Fisher information to measure the information content of local parameters, we retain local parameters with high Fisher values during the personalization process, which are considered informative, simultaneously prevent these parameters from noise perturbation. Secondly, we introduce an adaptive approach by applying differential constraint strategies to personalized parameters and shared parameters identified in the previous for better convergence. Our method boosts performance through flexible personalization while mitigating the slow convergence caused by clipping operations. Experimental results on CIFAR-10, FEMNIST and SVHN dataset demonstrate the effectiveness of our approach in achieving better performance and robustness against clipping, under personalized federated learning with differential privacy.

Visual intention understanding is to mine the potential and subjective intention behind the images, which includes the user's hidden emotions and perspectives. Due to the label ambiguity, this paper presents a novel learnable Hierarchical Label Embedding and Grouping (HLEG). It is featured in three aspects: 1) For effectively mining the underlying meaning of images, we build a hierarchical transformer structure to model the hierarchy of labels, formulating a multi-level classification scheme. 2) For the label ambiguity issue, we design a novel learnable label embedding with accumulative grouping integrated into the hierarchical structure, which does not require additional annotation. 3) For multi-level classification, we propose a “Hard-First” optimization strategy to adaptively adjust the classification optimization at different levels, avoiding over-classification of the coarse labels. HLEG enhances the F1 score (average +1.24%) and mAP (average +1.48%) on Intentonomy over prominent baseline models. Comprehensive experiments validate the superiority of our proposed method, achieving state-of-the-art performance under various settings. Code is available athttps://github.com/ShiQingHongYa/HLEG.

For the long-term person re-identification (ReID) task, pedestrians are likely to change clothes, which poses a key challenge in overcoming drastic appearance variations caused by these cloth changes. However, analyzing how cloth changes influence identity-invariant representation learning is difficult. In this context, varying cloth-changed samples are not adaptively utilized, and their effects on the resulting features are overshadowed. To address these limitations, this paper aims to estimate the effect of cloth-changing patterns at both the image and feature levels, presenting a Dual-Level Adaptive Weighting (DLAW) solution. Specifically, at the image level, we propose an adaptive mining strategy to locate the cloth-changed regions for each identity. This strategy highlights the informative areas that have undergone changes, enhancing robustness against cloth variations. At the feature level, we estimate the degree of cloth-changing by modeling the correlation of part-level features and re-weighting identity-invariant feature components. This further eliminates the effects of cloth variations at the semantic body part level. Extensive experiments demonstrate that our method achieves promising performance on several cloth-changing datasets. Code and models are available at https: //github.com/fountaindream/DLAW.

Daytime visible modality (RGB) and night-time infrared (IR) modality person re-identification (VI-ReID) is a challenging cross-modality pedestrian retrieval problem. However, training a cross-modality ReID model requires plenty of cross-modality (visible-infrared) identity labels that are more expensive than single-modality person ReID. To alleviate this issue, this paper studies unsupervised domain adaptive visible infrared person re-identification (UDA-VI-ReID) task without the reliance on any cross-modality annotation. To transfer learned knowledge from the labelled visible source domain to the unlabelled visible-infrared target domain, we propose a Translation, Association and Augmentation (TAA) framework. Specifically, the modality translator is firstly utilized to transfer visible image to infrared image, formulating generated visible-infrared image pairs for cross-modality supervised training. A Robust Association and Mutual Learning (RAML) module is then designed to exploit the underlying relations between visible and infrared modalities for label noise modeling. Moreover, a Translation Supervision and Feature Augmentation (TSFA) module is designed to enhance the discriminability by enriching the supervision with feature augmentation and modality translation. The extensive experimental results demonstrate that our method significantly outperforms current state-of-the-art unsupervised methods under various settings, and even surpasses some supervised counterparts, providing a powerful baseline for UDA-VI-ReID.

Composing Text and Image to Image Retrieval (CTI-IR) aims at finding the target image, which matches the query image visually along with the query text semantically. However, existing works ignore the fact that the reference text usually serves multiple functions, e.g., modification and auxiliary. To address this issue, we put forth a unified solution, namely Hierarchical Aggregation Transformer incorporated with Cross Relation Network (CRN). CRN unifies modification and relevance manner in a single framework. This configuration shows broader applicability, enabling us to model both modification and auxiliary text or their combination in triplet relationships simultaneously. Specifically, CRN includes: 1) Cross Relation Network comprehensively captures the relationships of various composed retrieval scenarios caused by two different query text types, allowing a unified retrieval model to designate adaptive combination strategies for flexible applicability; 2) Hierarchical Aggregation Transformer aggregates top-down features with Multi-layer Perceptron (MLP) to overcome the limitations of edge information loss in a window-based multi-stage Transformer. Extensive experiments demonstrate the superiority of the proposed CRN over all three fashion-domain datasets. Code is available at github.com/yan9qu/crn.

Visual intention understanding is the task of exploring the potential and underlying meaning expressed in images. Simply modeling the objects or backgrounds within the image content leads to unavoidable comprehension bias. To alleviate this problem, this paper proposes a Cross-modality Pyramid Alignment with Dynamic optimization (CPAD) to enhance the global understanding of visual intention with hierarchical modeling. The core idea is to exploit the hierarchical relationship between visual content and textual intention labels. For visual hierarchy, we formulate the visual intention understanding task as a hierarchical classification problem, capturing multiple granular features in different layers, which corresponds to hierarchical intention labels. For textual hierarchy, we directly extract the semantic representation from intention labels at different levels, which supplements the visual content modeling without extra manual annotations. Moreover, to further narrow the domain gap between different modalities, a cross-modality pyramid alignment module is designed to dynamically optimize the performance of visual intention understanding in a joint learning manner. Comprehensive experiments intuitively demonstrate the superiority of our proposed method, outperforming existing visual intention understanding methods.

Coronavirus disease 2019 (COVID-19) has become a severe global pandemic. Accurate pneumonia infection segmentation is important for assisting doctors in diagnosing COVID-19. Deep learning-based methods can be developed for automatic segmentation, but the lack of large-scale well-annotated COVID-19 training datasets may hinder their performance. Semi-supervised segmentation is a promising solution which explores large amounts of unlabelled data, while most existing methods focus on pseudo-label refinement. In this paper, we propose a new perspective on semi-supervised learning for COVID-19 pneumonia infection segmentation, namely pseudo-label guided image synthesis. The main idea is to keep the pseudo-labels and synthesize new images to match them. The synthetic image has the same COVID-19 infected regions as indicated in the pseudo-label, and the reference style extracted from the style code pool is added to make it more realistic. We introduce two representative methods by incorporating the synthetic images into model training, including single-stage Synthesis-Assisted Cross Pseudo Supervision (SA-CPS) and multi-stage Synthesis-Assisted Self-Training (SA-ST), which can work individually as well as cooperatively. Synthesis-assisted methods expand the training data with high-quality synthetic data, thus improving the segmentation performance. Extensive experiments on two COVID-19 CT datasets for segmenting the infections demonstrate our method is superior to existing schemes for semi-supervised segmentation, and achieves the state-of-the-art performance on both datasets. Code is available at: https://github.com/FeiLyu/SASSL.

Few-shot learning is a tough topic to solve since obtaining a large number of training samples in real applications is challenging. It has attracted increasing attention recently. Meta-learning is a prominent way to address this issue, intending to adapt predictors as base-learners to new tasks swiftly. However, a key challenge of meta-learning is its lack of expressive capacity, which stems from the difficulty of extracting general information from a small number of training samples. As a result, the generalizability of meta-learners trained from high-dimensional parameter spaces is frequently limited. To learn a better representation, we propose a graph complemented latent representation (GCLR) network for few-shot image classification. In particular, we embed the representation into a latent space, in which the latent codes are reconstructed using variational information to enrich the representation. In this way, the latent representation can achieve better generalizability. Another benefit is that, because the latent space is formed using variational inference, it cooperates well with various base-learners, boosting robustness. To make full use of the relation between samples in each category, a graph neural network (GNN) is also incorporated to improve relation mining. Consequently, our end-to-end framework delivers competitive performance on three few-shot learning benchmarks for image classification.

Single sample per person face recognition (SSPP FR) is one of the most challenging problems in FR due to the extreme lack of enrolment data. To date, the most popular SSPP FR methods are the generic learning methods, which recognize query face images based on the so-called prototype plus variation (i.e., P+V) model. However, the classic P+V model suffers from two major limitations: 1) it linearly combines the prototype and variation images in the observational pixel-spatial space and cannot generalize to multiple nonlinear variations, e.g., poses, which are common in face images and 2) it would be severely impaired once the enrolment face images are contaminated by nuisance variations. To address the two limitations, it is desirable to disentangle the prototype and variation in a latent feature space and to manipulate the images in a semantic manner. To this end, we propose a novel disentangled prototype plus variation model, dubbed DisP+V, which consists of an encoder-decoder generator and two discriminators. The generator and discriminators play two adversarial games such that the generator nonlinearly encodes the images into a latent semantic space, where the more discriminative prototype feature and the less discriminative variation feature are disentangled. Meanwhile, the prototype and variation features can guide the generator to generate an identity-preserved prototype and the corresponding variation, respectively. Experiments on various real-world face datasets demonstrate the superiority of our DisP+V model over the classic P+V model for SSPP FR. Furthermore, DisP+V demonstrates its unique characteristics in both prototype recovery and face editing/interpolation.

Model heterogeneous federated learning is a challenging task since each client independently designs its own model. Due to the annotation difficulty and free-riding par-ticipant issue, the local client usually contains unavoidable and varying noises, which cannot be effectively addressed by existing algorithms. This paper starts the first attempt to study a new and challenging robust federated learning problem with noisy and heterogeneous clients. We present a novel solution RHFL (Robust Heterogeneous Federated Learning), which simultaneously handles the label noise and performs federated learning in a single framework. It is featured in three aspects: (1) For the communication be-tween heterogeneous models, we directly align the models feedback by utilizing public data, which does not require additional shared global models for collaboration. (2) For internal label noise, we apply a robust noise-tolerant loss function to reduce the negative effects. (3) For challenging noisy feedback from other participants, we design a novel client confidence re-weighting scheme, which adaptively as-signs corresponding weights to each client in the collabo-rative learning stage. Extensive experiments validate the effectiveness of our approach in reducing the negative ef-fects of different noise rates/types under both model ho-mogeneous and heterogeneous federated learning settings, consistently outperforming existing methods.

Federated learning has emerged as an important distributed learning paradigm, which normally involves collaborative updating with others and local updating on private data. However, heterogeneity problem and catastrophic forgetting bring distinctive challenges. First, due to non-i.i.d (identically and independently distributed) data and heterogeneous architectures, models suffer performance degradation on other domains and communication barrier with participants models. Second, in local updating, model is separately optimized on private data, which is prone to overfit current data distribution and forgets previously acquired knowledge, resulting in catastrophic forgetting. In this work, we propose FCCL (Federated CrossCorrelation and Continual Learning). For heterogeneity problem, FCCL leverages unlabeled public data for communication and construct cross-correlation matrix to learn a generalizable representation under domain shift. Mean- while, for catastrophic forgetting, FCCL utilizes knowledge distillation in local updating, providing inter and intra domain information without leaking privacy. Empirical results on various image classification tasks demonstrate the effectiveness of our method and the efficiency of modules.

Deep hashing has gained growing popularity in approximate nearest neighbor search for large-scale image retrieval. Until now, the deep hashing for the image retrieval community has been dominated by convolutional neural network architectures, e.g. Resnet [22]. In this paper, inspired by the recent advancements of vision transformers, we present Transhash, a pure transformer-based framework for deep hashing learning. Concretely, our framework is composed of two major modules: (1) Based onVision Transformer (ViT), we design a siamese Multi-Granular Vision Tansformer backbone (MGVT) for image feature extraction. To learn fine-grained features, we innovate a dual-stream multi-granular feature learning on top of the transformer to learn discriminative global and local features. (2) Besides, we adopt a Bayesian learning scheme with a dynamically constructed similarity matrix to learn compact binary hash codes. The entire framework is jointly trained in an end-to-end manner. To the best of our knowledge, this is the first work to tackle deep hashing learning problems without convolutional neural networks (CNNs). We perform comprehensive experiments on three widely-studied datasets: CIFAR-10, NUSWIDE and IMAGENET. The experiments have evidenced our superiority against the existing state-of-the-art deep hashing methods. Specifically, we achieve 8.2%, 2.6%, 12.7% performance gains in terms of average mAP for different hash bit lengths on three public datasets, respectively.

Recognizing a target of interest from the UAVs is much more challenging than the existing object re-identification tasks across multiple city cameras. The images taken by the UAVs usually suffer from significant size difference when generating the object bounding boxes and uncertain rotation variations. Existing methods are usually designed for city cameras, incapable of handing the rotation issue in UAV scenarios. A straightforward solution is to perform the image-level rotation augmentation, but it would cause loss of useful information when inputting the powerful vision transformer as patches. This motivates us to simulate the rotation operation at the patch feature level, proposing a novel rotation invariant vision transformer (RotTrans). This strategy builds on high-level features with the help of the specificity of the vision transformer structure, which enhances the robustness against large rotation differences. In addition, we design invariance constraint to establish the relationship between the original feature and the rotated features, achieving stronger rotation invariance. Our proposed transformer tested on the latest UAV datasets greatly outperforms the current state-of-the-arts, which is 5.9% and 4.8% higher than the highest mAP and Rank1. Notably, our model also performs competitively for the person re-identification task on traditional city cameras. In particular, our solution wins the first place in the UAV-based person re-recognition track in the Multi-Modal Video Reasoning and Analyzing Competition held in ICCV 2021. Code is available at https://github.com/whucsy/RotTrans.

Sketch-photo recognition is a cross-modal matching problem whose query sets are sketch images drawn by artists or amateurs. Due to the significant modality difference between the two modalities, it is challenging to extract discriminative modality-shared feature representations. Existing works focus on exploring modality-invariant features to discover shared embedding space. However, they discard modality-specific cues, resulting in information loss and diminished discriminatory power of features. This paper proposes a novel asymmetrical disentanglement and dynamic synthesis learning method in the transformer framework (SketchTrans) to handle modality discrepancy by combining modality-shared information with modality-specific information. Specifically, an asymmetrical disentanglement scheme is introduced to decompose the photo features into sketch-relevant and sketch-irrelevant cues while preserving the original sketch structure. Using the sketch-irrelevant cues, we further translate the sketch modality component to photo representation through knowledge transfer, obtaining cross-modality representations with information symmetry. Moreover, we propose a dynamic updatable auxiliary sketch (A-sketch) modality generated from the photo modality to guide the asymmetrical disentanglement in a single framework. Under a multi-modality joint learning framework, this auxiliary modality increases the diversity of training samples and narrows the cross-modality gap. We conduct extensive experiments on three fine-grained sketch-based retrieval datasets, i.e., PKU-Sketch, QMUL-ChairV2, and QMUL-ShoeV2, outperforming the state-of-the-arts under various metrics.

Federated learning has received increasing attention for its ability to collaborative learning without leaking privacy. Promising advances have been achieved under the assumption that participants share the same model structure. However, when participants independently customize their models, models suffer communication barriers, which leads the model heterogeneity problem. Moreover, in real scenarios, the data held by participants is often limited, making the local models trained only on private data present poor performance. Consequently, this paper studies a new challenging problem, namely few-shot model agnostic federated learning, where the local participants design their independent models from their limited private datasets. Considering the scarcity of the private data, we propose to utilize the abundant public available datasets for bridging the gap between local private participants. However, its usage also brings in two problems: inconsistent labels and large domain gap between the public and private datasets. To address these issues, this paper presents a novel framework with two main parts: 1) model agnostic federated learning, it performs public-private communication by unifying the model prediction outputs on the shared public datasets; 2) latent embedding adaptation, it addresses the domain gap with an adversarial learning scheme to discriminate the public and private domains. Together with theoretical generalization bound analysis, comprehensive experiments under various settings have verified our advantage over existing methods. It provides a simple but effective baseline for future advancement. The code is available at https://github.com/WenkeHuang/FSMAFL.

The robust and discriminative feature extraction is the key component in person re-identification (Re-ID). The major weakness of conventional convolution neural network (CNN) based methods is that they cannot extract long-range information from diverse parts, which can be alleviated by recently developed Transformers. Existing vision Transformers show their power on various vision tasks. However, they (i) cannot address translation problems and different viewpoints; (ii) cannot capture detailed features to discriminate people with a similar appearance. In this paper, we propose a powerful Re-ID baseline built on top of the pyramidal transformer with conv-patchify operation, termed PTCR, which inherits the advantages of both CNN and Transformer. The pyramidal structure captures multi-scale fine-grained features, while the conv-patchify enhances the robustness against translation. Moreover, we additionally design two novel modules to improve the robust feature learning. A Token Perception module augments the patch embeddings to enhance the robustness against perturbation and viewpoint changes, while the Auxiliary Embedding module integrates the auxiliary information (cam ID, pedestrian attributes, etc) to reduce feature bias caused by non-visual factors. Our method is validated through extensive experiments to show its superior performance with abundant ablation studies. Notably, without re-ranking, we achieve 98.0% Rank-1 on Market-1501 and 88.6% Rank-1 on MSMT17, significantly outperforming the counterparts. The code is available at: https://github.com/lihe404/PTCR

Color-guided depth super-resolution (DSR) is an encouraging paradigm that enhances a low-resolution (LR) depth map guided by an extra high-resolution (HR) RGB image from the same scene. Existing methods usually use interpolation to upscale the depth maps before feeding them into the network and transfer the high-frequency information extracted from HR RGB images to guide the reconstruction of depth maps. However, the extracted high-frequency information usually contains textures that are not present in depth maps in the existence of the cross-modality gap, and the noises would be fur- ther aggravated by interpolation due to the resolution gap between the RGB and depth images. To tackle these challenges, we propose a novel Symmetric Uncertainty-aware Feature Transmission (SUFT) for color-guided DSR. (1) For the resolution gap, SUFT builds an iterative up-and-down sampling pipeline, which makes depth features and RGB features spatially consistent while suppressing noise amplification and blurring by replacing common interpolated pre-upsampling. (2) For the cross-modality gap, we propose a novel Symmetric Uncertainty scheme to remove parts of RGB information harmful to the recovery of HR depth maps. Extensive experiments on benchmark datasets and challenging real-world settings suggest that our method achieves superior performance compared to state-of-the-art methods. Our code and models are available at https://github.com/ShiWuxuan/SUFT.

Audio-visual speaker diarization aims at detecting "who spoke when'' using both auditory and visual signals. Existing audio-visual diarization datasets are mainly focused on indoor environments like meeting rooms or news studios, which are quite different from in-the-wild videos in many scenarios such as movies, documentaries, and audience sitcoms. To develop diarization methods for these challenging videos, we create the AVA Audio-Visual Diarization (AVA-AVD) dataset. Our experiments demonstrate that adding AVA-AVD into training set can produce significantly better diarization models for in-the-wild videos despite that the data is relatively small. Moreover, this benchmark is challenging due to the diverse scenes, complicated acoustic conditions, and completely off-screen speakers. As a first step towards addressing the challenges, we design the Audio-Visual Relation Network (AVR-Net) which introduces a simple yet effective modality mask to capture discriminative information based on face visibility. Experiments show that our method not only can outperform state-of-the-art methods but is more robust as varying the ratio of off-screen speakers. Our data and code has been made publicly available at \textcolormagenta \urlhttps://github.com/showlab/AVA-AVD .

Visible infrared person re-identification (VI-ReID) aims at searching out the corresponding infrared (visible) images from a gallery set captured by other spectrum cameras. Recent works mainly focus on supervised VI-ReID methods that require plenty of cross-modality (visible-infrared) identity labels which are more expensive than the annotations in single-modality person ReID. For the unsupervised learning visible infrared re-identification (USL-VI-ReID), the large cross-modality discrepancies lead to difficulties in generating reliable cross-modality labels and learning modality-invariant features without any annotations. To address this problem, we propose a novel Augmented Dual-Contrastive Aggregation (ADCA) learning framework. Specifically, a dual-path contrastive learning framework with two modality-specific memories is proposed to learn the intra-modality person representation. To associate positive cross-modality identities, we design a cross-modality memory aggregation module with count priority to select highly associated positive samples, and aggregate their corresponding memory features at the cluster level, ensuring that the optimization is explicitly concentrated on the modality-irrelevant perspective. Extensive experiments demonstrate that our proposed ADCA significantly outperforms existing unsupervised methods under various settings, and even surpasses some supervised counterparts, facilitating VI-ReID to real-world deployment. Code is available at https://github.com/yangbincv/ADCA.

This paper studies a novel privacy-preserving anonymization problem for pedestrian images, which preserves personal identity information (PII) for authorized models and prevents PII from being recognized by third parties. Conventional anonymization methods unavoidably cause semantic information loss, leading to limited data utility. Besides, existing learned anonymization techniques, while retaining various identity-irrelevant utilities, will change the pedestrian identity, and thus are unsuitable for training robust re-identification models. To explore the privacy-utility trade-off for pedestrian images, we propose a joint learning reversible anonymization framework, which can reversibly generate full-body anonymous images with little performance drop on person re-identification tasks. The core idea is that we adopt desensitized images generated by conventional methods as the initial privacy-preserving supervision and jointly train an anonymization encoder with a recovery decoder and an identity-invariant model. We further propose a progressive training strategy to improve the performance, which iteratively upgrades the initial anonymization supervision. Experiments further demonstrate the effectiveness of our anonymized pedestrian images for privacy protection, which boosts the re-identification performance while preserving privacy. Code is available at https://github.com/whuzjw/privacy-reid.

Person re-identification (Re-ID) aims at retrieving a person of interest across multiple non-overlapping cameras. With the advancement of deep neural networks and increasing demand of intelligent video surveillance, it has gained significantly increased interest in the computer vision community. By dissecting the involved components in developing a person Re-ID system, we categorize it into the closed-world and open-world settings. The widely studied closed-world setting is usually applied under various research-oriented assumptions, and has achieved inspiring success using deep learning techniques on a number of datasets. We first conduct a comprehensive overview with in-depth analysis for closed-world person Re-ID from three different perspectives, including deep feature representation learning, deep metric learning and ranking optimization. With the performance saturation under closed-world setting, the research focus for person Re-ID has recently shifted to the open-world setting, facing more challenging issues. This setting is closer to practical applications under specific scenarios. We summarize the open-world Re-ID in terms of five different aspects. By analyzing the advantages of existing methods, we design a powerful AGW baseline, achieving state-of-the-art or at least comparable performance on twelve datasets for four different Re-ID tasks. Meanwhile, we introduce a new evaluation metric (mINP) for person Re-ID, indicating the cost for finding all the correct matches, which provides an additional criteria to evaluate the Re-ID system for real applications. Finally, some important yet under-investigated open issues are discussed.

Deep embedding learning plays a key role in learning discriminative feature representations, where the visually similar samples are pulled closer and dissimilar samples are pushed away in the low-dimensional embedding space. This paper studies the unsupervised embedding learning problem by learning such a representation without using any category labels. This task faces two primary challenges: mining reliable positive supervision from highly similar fine-grained classes, and generalizing to unseen testing categories. To approximate the positive concentration and negative separation properties in category-wise supervised learning, we introduce a data augmentation invariant and instance spreading feature using the instance-wise supervision. We also design two novel domain-agnostic augmentation strategies to further extend the supervision in feature space, which simulates the large batch training using a small batch size and the augmented features. To learn such a representation, we propose a novel instance-wise softmax embedding, which directly perform the optimization over the augmented instance features with the binary discrmination softmax encoding. It significantly accelerates the learning speed with much higher accuracy than existing methods, under both seen and unseen testing categories. The unsupervised embedding performs well even without pre-trained network over samples from fine-grained categories. We also develop a variant using category-wise supervision, namely category-wise softmax embedding, which achieves competitive performance over the state-of-of-the-arts, without using any auxiliary information or restrict sample mining.

Video-based person re-identification (ReID) matches the same people across the video sequences with rich spatial and temporal information in complex scenes. It is highly challenging to capture discriminative information when occlusions and pose variations exist between frames. A key solution to this problem rests on extracting the temporal invariant features of video sequences. In this paper, we propose a novel method for discovering temporal coherence by designing a region-level saliency and granularity mining network (SGMN). Firstly, to address the varying noisy frame problem, we design a temporal spatial-relation module (TSRM) to locate frame-level salient regions, adaptively modeling the temporal relations on spatial dimension through a probe-buffer mechanism. It avoids the information redundancy between frames and captures the informative cues of each frame. Secondly, a temporal channel-relation module (TCRM) is proposed to further mine the small granularity information of each frame, which is complementary to TSRM by concentrating on discriminative small-scale regions. TCRM exploits a one-and-rest difference relation on channel dimension to enhance the granularity features, leading to stronger robustness against misalignments. Finally, we evaluate our SGMN with four representative video-based datasets, including iLIDS-VID, MARS, DukeMTMC-VideoReID, and LS-VID, and the results indicate the effectiveness of the proposed method.

Visible-infrared person re-identification (VI-ReID) is an emerging and challenging cross-modality image matching problem because of the explosive surveillance data in night-time surveillance applications. To handle the large modality gap, various generative adversarial network models have been developed to eliminate the cross-modality variations based on a cross-modal image generation framework. However, the lack of point-wise cross-modality ground-truths makes it extremely challenging to learn such a cross-modal image generator. To address these problems, we learn the correspondence between single-channel infrared images and three-channel visible images by generating intermediate grayscale images as auxiliary information to colorize the single-modality infrared images. We propose a grayscale enhancement colorization network (GECNet) to bridge the modality gap by retaining the structure of the colored image which contains rich information. To simulate the infrared-to-visible transformation, the point-wise transformed grayscale images greatly enhance the colorization process. Our experiments conducted on two visible-infrared cross-modality person re-identification datasets demonstrate the superiority of the proposed method over the state-of-the-arts.

Recently, the correlation filter (CF) and Siamese network have become the two most popular frameworks in object tracking. Existing CF trackers, however, are limited by feature learning and context usage, making them sensitive to boundary effects. In contrast, Siamese trackers can easily suffer from the interference of semantic distractors. To address the above problems, we propose an end-to-end target-insight correlation network (TICNet) for object tracking, which aims at breaking the above limitations on top of a unified network. TICNet is an asymmetric dual-branch network involving a target-background awareness model (TBAM), a spatial-channel attention network (SCAN), and a distractor-aware filter (DAF) for end-to-end learning. Specifically, TBAM aims to distinguish a target from the background in the pixel level, yielding a target likelihood map based on color statistics to mine distractors for DAF learning. SCAN consists of a basic convolutional network, a channel-attention network, and a spatial-attention network, aiming to generate attentive weights to enhance the representation learning of the tracker. Especially, we formulate a differentiable DAF and employ it as a learnable layer in the network, thus helping suppress distracting regions in the background. During testing, DAF, together with TBAM, yields a response map for the final target estimation. Extensive experiments on seven benchmarks demonstrate that TICNet outperforms the state-of-the-art methods while running at real-time speed.

Medical time series of laboratory tests has been collected in electronic health records (EHRs) in many countries. Machine-learning algorithms have been proposed to analyze the condition of patients using these medical records. However, medical time series may be recorded using different laboratory parameters in different datasets. This results in the failure of applying a pretrained model on a test dataset containing a time series of different laboratory parameters. This article proposes to solve this problem with an unsupervised time-series adaptation method that generates time series across laboratory parameters. Specifically, a medical time-series generation network with similarity distillation is developed to reduce the domain gap caused by the difference in laboratory parameters. The relations of different laboratory parameters are analyzed, and the similarity information is distilled to guide the generation of target-domain specific laboratory parameters. To further improve the performance in cross-domain medical applications, a missingness-aware feature extraction network is proposed, where the missingness patterns reflect the health conditions and, thus, serve as auxiliary features for medical analysis. In addition, we also introduce domain-adversarial networks in both feature level and time-series level to enhance the adaptation across domains. Experimental results show that the proposed method achieves good performance on both private and publicly available medical datasets. Ablation studies and distribution visualization are provided to further analyze the properties of the proposed method.

Most existing works solve the video-based person re-identification (re-ID) problem by computing the representation of each frame independently and finally aggregate the frame-level features. However, these methods often suffer from the challenging factors in videos, such as serious occlusion, background clutter and pose variation. To address these issues, we propose a novel multi-level Context-aware Part Attention (CPA) model to learn discriminative and robust local part features. It is featured in two aspects: 1) the context-aware part attention module improves the robustness by capturing the global relationship among different body parts across different video frames, and 2) the attention module is further extended to multi-level attention mechanism which enhances the discriminability by simultaneously considering low- to high-level features in different convolutional layers. In addition, we propose a novel multi-head collaborative training scheme to improve the performance, which is collaboratively supervised by multiple heads with the same structure but different parameters. It contains two consistency regularization terms, which consider both multi-head and multi-frame consistency to achieve better results. The multi-level CPA model is designed for feature extraction, while the multi-head collaborative training scheme is designed for classifier supervision. They jointly improve our re-ID model from two complementary directions. Extensive experiments demonstrate that the proposed method achieves much better or at least comparable performance compared to the state-of-the-art on four video re-ID datasets.

Matching the daytime visible and nighttime infrared person images, namely visible infrared person re-identification (VI-ReID), is a challenging cross-modality retrieval problem. Due to the difficulty of data collection and annotation in nighttime surveillance, VI-ReID usually suffers from noise problems, making it challenging to directly learn part discriminative features. In order to improve the discriminability and enhance the robustness against noisy images, this paper proposes a novel dynamic tri-level relation mining (DTRM) framework by simultaneously exploring channel-level, part-level intra-modality, and graph-level cross-modality relation cues. To address the misalignment within the person images, we design an intra-modality weighted-part attention (IWPA) to construct part-aggregated representation. It adaptively integrates the body part relation into the local feature learning with a residual batch normalization (RBN) connection scheme. Besides, a cross-modality graph structured attention (CGSA) is incorporated to improve the global feature learning by utilizing the contextual relation between images from two modalities. This module reduces the negative effects of noisy images. To seamlessly integrate two components, a parameter-free dynamic aggregation strategy is designed in a progressive joint learning manner. To further improve the performance, we additionally design a simple yet effective channel-level learning strategy by exploiting the rich channel information of visible images, which significantly reinforces the performance without modifying the network structure or changing the training process. Extensive experiments on two visible infrared re-identification datasets have verified the effectiveness under various settings. Code is available at:https://github.com/mangye16/DDAG

Visible-infrared person re-identification (VI-ReID) is a cross-modality retrieval problem, which aims at matching the same pedestrian between the visible and infrared cameras. Due to the existence of pose variation, occlusion, and huge visual differences between the two modalities, previous studies mainly focus on learning image-level shared features. Since they usually learn a global representation or extract uniformly divided part features, these methods are sensitive to misalignments. In this paper, we propose a structure-aware positional transformer (SPOT) network to learn semantic-aware sharable modality features by utilizing the structural and positional information. It consists of two main components: attended structure representation (ASR) and transformer-based part interaction (TPI). Specifically, ASR models the modality-invariant structure feature for each modality and dynamically selects the discriminative appearance regions under the guidance of the structure information. TPI mines the part-level appearance and position relations with a transformer to learn discriminative part-level modality features. With a weighted combination of ASR and TPI, the proposed SPOT explores the rich contextual and structural information, effectively reducing cross-modality difference and enhancing the robustness against misalignments. Extensive experiments indicate that SPOT is superior to the state-of-the-art methods on two cross-modal datasets. Notably, the Rank-1/mAP value on the SYSU-MM01 dataset has improved by 8.43%/6.80%.

This paper studies the challenging person re-identification (Re-ID) task under the cloth-changing scenario, where the same identity (ID) suffers from uncertain cloth changes. To learn cloth- and ID-invariant features, it is crucial to collect abundant training data with varying clothes, which is difficult in practice. To alleviate the reliance on rich data collection, we reinforce the feature learning process by designing powerful complementary data augmentation strategies, including positive and negative data augmentation. Specifically, the positive augmentation fulfills the ID space by randomly patching the person images with different clothes, simulating rich appearance to enhance the robustness against clothes variations. For negative augmentation, its basic idea is to randomly generate out-of-distribution synthetic samples by combining various appearance and posture factors from real samples. The designed strategies seamlessly reinforce the feature learning without additional information introduction. Extensive experiments conducted on both cloth-changing and -unchanging tasks demonstrate the superiority of our proposed method, consistently improving the accuracy over various baselines.

Existing person re-identification (Re-ID) methods usually rely heavily on large-scale thoroughly annotated training data. However, label noise is unavoidable due to inaccurate person detection results or annotation errors in real scenes. It is extremely challenging to learn a robust Re-ID model with label noise since each identity has very limited annotated training samples. To avoid fitting to the noisy labels, we propose to learn a prefatory model using a large learning rate at the early stage with a self-label refining strategy, in which the labels and network are jointly optimized. To further enhance the robustness, we introduce an online co-refining (CORE) framework with dynamic mutual learning, where networks and label predictions are online optimized collaboratively by distilling the knowledge from other peer networks. Moreover, it also reduces the negative impact of noisy labels using a favorable selective consistency strategy. CORE has two primary advantages: it is robust to different noise types and unknown noise ratios; it can be easily trained without much additional effort on the architecture design. Extensive experiments on Re-ID and image classification demonstrate that CORE outperforms its counterparts by a large margin under both practical and simulated noise settings. Notably, it also improves the state-of-the-art unsupervised Re-ID performance under standard settings. Code is available at https://github.com/mangye16/ReID-Label-Noise.

Automatic liver tumor segmentation could offer assistance to radiologists in liver tumor diagnosis, and its performance has been significantly improved by recent deep learning based methods. These methods rely on large-scale well-annotated training datasets, but collecting such datasets is time-consuming and labor-intensive, which could hinder their performance in practical situations. Learning from synthetic data is an encouraging solution to address this problem. In our task, synthetic tumors can be injected to healthy images to form training pairs. However, directly applying the model trained using the synthetic tumor images on real test images performs poorly due to the domain shift problem. In this paper, we propose a novel approach, namely Synthetic-to-Real Test-Time Training (SR-TTT), to reduce the domain gap between synthetic training images and real test images. Specifically, we add a self-supervised auxiliary task, i.e., two-step reconstruction, which takes the output of the main segmentation task as its input to build an explicit connection between these two tasks. Moreover, we design a scheduled mixture strategy to avoid error accumulation and bias explosion in the training process. During test time, we adapt the segmentation model to each test image with self-supervision from the auxiliary task so as to improve the inference performance. The proposed method is extensively evaluated on two public datasets for liver tumor segmentation. The experimental results demonstrate that our proposed SR-TTT can effectively mitigate the synthetic-to-real domain shift problem in the liver tumor segmentation task, and is superior to existing state-of-the-art approaches.

Cross-modality person re-identification is a challenging task due to large cross-modality discrepancy and intramodality variations. Currently, most existing methods focus on learning modality-specific or modality-shareable features by using the identity supervision or modality label. Different from existing methods, this paper presents a novel Modality Confusion Learning Network (MCLNet). Its basic idea is to confuse two modalities, ensuring that the optimization is explicitly concentrated on the modality-irrelevant perspective. Specifically, MCLNet is designed to learn modality-invariant features by simultaneously minimizing inter-modality discrepancy while maximizing cross-modality similarity among instances in a single framework. Furthermore, an identity-aware marginal center aggregation strategy is introduced to extract the centralization features, while keeping diversity with a marginal constraint. Finally, we design a camera-aware learning scheme to enrich the discriminability. Extensive experiments on SYSU-MM01 and RegDB datasets show that MCLNet outperforms the state-of-the-art by a large margin. On the large-scale SYSU-MM01 dataset, our model can achieve 65.40 % and 61.98 % in terms of Rank-1 accuracy and mAP value.

This paper introduces a powerful channel augmented joint learning strategy for the visible-infrared recognition problem. For data augmentation, most existing methods directly adopt the standard operations designed for single-modality visible images, and thus do not fully consider the imagery properties in visible to infrared matching. Our basic idea is to homogenously generate color-irrelevant images by randomly exchanging the color channels. It can be seamlessly integrated into existing augmentation operations without modifying the network, consistently improving the robustness against color variations. Incorporated with a random erasing strategy, it further greatly enriches the diversity by simulating random occlusions. For cross-modality metric learning, we design an enhanced channel-mixed learning strategy to simultaneously handle the intra-and cross-modality variations with squared difference for stronger discriminability. Besides, a channel-augmented joint learning strategy is further developed to explicitly optimize the outputs of augmented images. Extensive experiments with insightful analysis on two visible-infrared recognition tasks show that the proposed strategies consistently improve the accuracy. Without auxiliary information, it improves the state-of-the-art Rank-1/mAP by 14.59%/13.00% on the large-scale SYSU-MM01 dataset.

Single sample per person face recognition (SSPP FR) is one of the most challenging problems in FR due to the extreme lack of enrolment data. State-of-the-art SSPP FR methods are based on the prototype plus variation (i.e., P+V) model. However, the classic P+V model has two major limitations: 1) It is a linear model and cannot generalize many non-linear variations; 2) It can be severely impaired once the enrolment face images are contaminated with variations. To this end, we propose a novel disentangled prototype plus variation model, dubbed DisP+V, to tackle such limitations. DisP+V consists of an encoder-decoder structural generator and two discriminators. The generator and discriminators play two adversarial games such that the generator nonlinearly encodes the images into a latent semantic space, where the more discriminative prototype feature and the less discriminative variation feature are disentangled. Meanwhile, the prototype and variation features in the latent space can guide the generator to generate an identity-preserved prototype and the corresponding variation, respectively. Experiments on various real-world face datasets demonstrate the superiority of our DisP+V model over the classic P+V model for SSPP FR. Furthermore, DisP+V demonstrates its unique characteristics in the challenging prototype recovery task.

Image-text retrieval is one of the most common tasks in multimodal retrieval. It suffers from the problem of information imbalance between modalities, which is so-called modality gap. It remains challenging because prior methods cannot bridge the gap reasonably. With the help of scene graph, we start by designing an auxiliary bi-level graph representation (ABGR) pipeline that can fully mine the potential information and reduce the information redundancy. By doing so, each modality will be represented by lexical word graph that carries the main content of the information. Specifically, we design a graph feature enhancement (GFE) module to embed the graph-structured information in a common subspace while exploring the relationship between lexical words. As a result, a better representation for both image and text can be obtained, which helps us to evaluate the similarity between images and texts more reasonably. Experimental results conducted on two benchmark datasets Flickr30K and MS-COCO demonstrate the effectiveness of our proposed model for cross-modal retrieval task.

Due to the dynamic health status of patients and discrepant stability of physiological variables, health data often presents as irregular multi-rate multivariate time series (IMR-MTS) with significantly varying sampling rates. Existing methods mainly study changes of IMR-MTS values in the time domain, without considering their different dominant frequencies and varying data quality. Hence, we propose a novel Cooperative Joint Attentive Network (CJANet) to analyze IMR-MTS in frequency domain, which adaptively handling discrepant dominant frequencies while tackling diverse data qualities caused by irregular sampling. In particular, novel dual-channel joint attention is designed to jointly identify important magnitude and phase signals while detecting their dominant frequencies, automatically enlarging the positive influence of key variables and frequencies. Furthermore, a new cooperative learning module is introduced to enhance information exchange between magnitude and phase channels, effectively integrating global signals to optimize the network. A frequency-aware fusion strategy is finally designed to aggregate the learned features. Extensive experimental results on real-world medical datasets indicate that CJANet significantly outperforms existing methods and provides highly interpretable results.

Multi-person action forecasting is an emerging task and a pivotal step towards video understanding. The major challenge lies in estimating a distribution characterizing the upcoming actions of all individuals in the scene. The state-of-the-art solutions attempt to solve this problem via a step-by-step prediction procedure. However, they are not adequate to address some particular limitations, such as the compounding errors, the innate uncertainty of the future and the spatio-temporal contexts. To handle the multi-person action forecasting challenges, we put forth a novel imitative learning framework upon the basis of inverse reinforcement learning. Specifically, we aim to learn a policy to model the aforementioned distribution up to a coming horizon through an objective that naturally solves the compounding errors. Such a policy is able to explore multiple plausible futures via extrapolating a series of latent variables and taking them into account to generate predictions. The impacts of these latent variables are further investigated by optimizing the directed information. Moreover, we reason the spatial context along with the temporal cue in a single pass with the usage of graph structural data. The experimental outcomes on two large-scale datasets reveal that our approach yields considerable improvements in terms of both diversity and quality with respect to recent leading studies.

The aim of person re-identification (Re-ID) is retrieving a person of interest across multiple non-overlapping cameras. Re-ID has gained significantly increased advancement in recent years. However, real data annotation is costly and model generalization ability is hindered by the lack of large-scale and diverse data. To address this problem, we propose a Weather Person pipeline that can generate a synthesized Re-ID dataset with different weather, scenes, and natural lighting conditions automatically. The pipeline is built on the top of a game engine which contains a digital city, weather and lighting simulation system, and various character models with manifold dressing. To train a generalizable Re-ID model from the large-scale virtual WePerson dataset, we design an adaptive sample selection strategy to close the domain gap and avoid redundancy. We also design an informative sampling method for a mini-batch sampler to accelerate the learning process. In addition, an efficient training method is introduced by adopting instance normalization to capture identity invariant components from various appearances. We evaluate our pipeline using direct transfer on 3 widely-used real-world benchmarks, achieving competitive performance without any real-world image training. This dataset starts the attempt to evaluate diverse environmental factors in a controllable virtual engine, which provides important guidance for future generalizable Re-ID model design. Notably, we improve the current state-of-the-art accuracy from 38.5% to 46.4% on the challenging MSMT17 dataset. Dataset and code are available at https://github.com/lihe404/WePerson https://github.com/lihe404/WePerson.

OBJECTIVE: Accurate risk prediction is important for evaluating early medical treatment effects and improving health care quality. Existing methods are usually designed for dynamic medical data, which require long-term observations. Meanwhile, important personalized static information is ignored due to the underlying uncertainty and unquantifiable ambiguity. It is urgent to develop an early risk prediction method that can adaptively integrate both static and dynamic health data. MATERIALS AND METHODS: Data were from 6367 patients with Peptic Ulcer Bleeding between 2007 and 2016. This article develops a novel End-to-end Importance-Aware Personalized Deep Learning Approach (eiPDLA) to achieve accurate early clinical risk prediction. Specifically, eiPDLA introduces a long short-term memory with temporal attention to learn sequential dependencies from time-stamped records and simultaneously incorporating a residual network with correlation attention to capture their influencing relationship with static medical data. Furthermore, a new multi-residual multi-scale network with the importance-aware mechanism is designed to adaptively fuse the learned multisource features, automatically assigning larger weights to important features while weakening the influence of less important features. RESULTS: Extensive experimental results on a real-world dataset illustrate that our method significantly outperforms the state-of-the-arts for early risk prediction under various settings (eg, achieving an AUC score of 0.944 at 1 year ahead of risk prediction). Case studies indicate that the achieved prediction results are highly interpretable. CONCLUSION: These results reflect the importance of combining static and dynamic health data, mining their influencing relationship, and incorporating the importance-aware mechanism to automatically identify important features. The achieved accurate early risk prediction results save precious time for doctors to timely design effective treatments and improve clinical outcomes.

Matching person images between the daytime visible modality and night-time infrared modality (VI-ReID) is a challenging cross-modality pedestrian retrieval problem. Existing methods usually learn the multi-modality features in raw image, ignoring the image-level discrepancy. Some methods apply GAN technique to generate the cross-modality images, but it destroys the local structure and introduces unavoidable noise. In this paper, we propose a Homogeneous Augmented Tri-Modal (HAT) learning method for VI-ReID, where an auxiliary grayscale modality is generated from their homogeneous visible images, without additional training process. It preserves the structure information of visible images and approximates the image style of infrared modality. Learning with the grayscale visible images enforces the network to mine structure relations across multiple modalities, making it robust to color variations. Specifically, we solve the tri-modal feature learning from both multi-modal classification and multi-view retrieval perspectives. For multi-modal classification, we learn a multi-modality sharing identity classifier with a parameter-sharing network, trained with a homogeneous and heterogeneous identification loss. For multi-view retrieval, we develop a weighted tri-directional ranking loss to optimize the relative distance across multiple modalities. Incorporated with two invariant regularizers, HAT simultaneously minimizes multiple modality variations. In-depth analysis demonstrates the homogeneous grayscale augmentation significantly outperforms the current state-of-the-art by a large margin.

Influenced by the dynamic changes in the severity of illness, patients usually take examinations in hospitals irregularly, producing a large volume of irregular medical time-series data. Performing diagnosis prediction from the irregular medical time series is challenging because the intervals between consecutive records significantly vary along time. Existing methods often handle this problem by generating regular time series from the irregular medical records without considering the uncertainty in the generated data, induced by the varying intervals. Thus, a novel Uncertainty-Aware Convolutional Recurrent Neural Network (UA-CRNN) is proposed in this article, which introduces the uncertainty information in the generated data to boost the risk prediction. To tackle the complex medical time series with subseries of different frequencies, the uncertainty information is further incorporated into the subseries level rather than the whole sequence to seamlessly adjust different time intervals. Specifically, a hierarchical uncertainty-aware decomposition layer (UADL) is designed to adaptively decompose time series into different subseries and assign them proper weights in accordance with their reliabilities. Meanwhile, an Explainable UA-CRNN (eUA-CRNN) is proposed to exploit filters with different passbands to ensure the unity of components in each subseries and the diversity of components in different subseries. Furthermore, eUA-CRNN incorporates with an uncertainty-aware attention module to learn attention weights from the uncertainty information, providing the explainable prediction results. The extensive experimental results on three real-world medical data sets illustrate the superiority of the proposed method compared with the state-of-the-art methods.

Due to the discrepancy of diseases and symptoms, patients usually visit hospitals irregularly and different physiological variables are examined at each visit, producing large amounts of irregular multivariate time series (IMTS) data with missing values and varying intervals. Existing methods process IMTS into regular data so that standard machine learning models can be employed. However, time intervals are usually determined by the status of patients, while missing values are caused by changes in symptoms. Therefore, we propose a novel end-to-end Dual-Attention Time-Aware Gated Recurrent Unit (DATA-GRU) for IMTS to predict the mortality risk of patients. In particular, DATA-GRU is able to: 1) preserve the informative varying intervals by introducing a time-aware structure to directly adjust the influence of the previous status in coordination with the elapsed time, and 2) tackle missing values by proposing a novel dual-attention structure to jointly consider data-quality and medical-knowledge. A novel unreliability-aware attention mechanism is designed to handle the diversity in the reliability of different data, while a new symptom-aware attention mechanism is proposed to extract medical reasons from original clinical records. Extensive experimental results on two real-world datasets demonstrate that DATA-GRU can significantly outperform state-of-the-art methods and provide meaningful clinical interpretation.

Unsupervised embedding learning aims at extracting low-dimensional visually meaningful representations from large-scale unlabeled images, which can then be directly used for similarity-based search. This task faces two major challenges: 1) mining positive supervision from highly similar fine-grained classes and 2) generating to unseen testing categories. To tackle these issues, this paper proposes a probabilistic structural latent representation (PSLR), which incorporates an adaptable softmax embedding to approximate the positive concentrated and negative instance separated properties in the graph latent space. It improves the discriminability by enlarging the positive/negative difference without introducing any additional computational cost while maintaining high learning efficiency. To address the limited supervision using data augmentation, a smooth variational reconstruction loss is introduced by modeling the intra-instance variance, which improves the robustness. Extensive experiments demonstrate the superiority of PSLR over state-of-the-art unsupervised methods on both seen and unseen categories with cosine similarity. Code is available at https://github.com/mangye16/PSLR.

RGB-Infrared person re-identification (RGB-IR Re-ID) aims to matching persons from heterogeneous images captured by visible and thermal cameras, which is of great significance in the surveillance system under poor light conditions. Facing great challenges in complex variances including conventional single-modality and additional inter-modality discrepancies, most of the existing RGB-IR Re-ID methods propose to impose constraints in image level, feature level or a hybrid of both. Despite better performance of hybrid constraints, they are usually implemented with heavy network architecture. As a matter of fact, previous efforts contribute more as pioneering works in new cross-modal Re-ID area while leaving large space for improvement. This can be mainly attributed to: (a) lack of abundant person image pairs from different modalities for training, and (b) scarcity of salient modality-invariant features especially on coarse representations for effective matching. To address these issues, a novel Multi-Scale Part-Aware Cascading framework (MSPAC) is formulated by aggregating multi-scale fine-grained features from part to global in a cascading manner, which results in a unified representation containing rich and enhanced semantic features. Furthermore, a marginal exponential center (MeCen) loss is introduced to jointly eliminate mixed variances from intra- and inter-modal examples. Cross-modality correlations can thus be efficiently explored on salient features for distinctive modality-invariant feature learning. Extensive experiments are conducted to demonstrate that the proposed method outperforms all the state-of-the-art by a large margin.

Person re-identification (re-ID) has been a popular topic in computer vision and pattern recognition communities for a decade. Several important milestones such as metric-based and deeply-learned re-ID in recent years have promoted this topic. However, most existing re-ID works are designed for closed-world scenarios rather than realistic open-world settings, which limits the practical application of the re-ID technique. On one hand, the performance of the latest re-ID methods has surpassed the human-level performance on several commonly used benchmarks (e.g., Market1501 and CUHK03), which are collected from closed-world scenarios. On the other hand, open-world tasks that are less developed and more challenging have received increasing attention in the re-ID community. Therefore, this paper starts the first attempt to analyze the trends of open-world re-ID and summarizes them from both narrow and generalized perspectives. In the narrow perspective, open-world re-ID is regarded as person verification (i.e., open-set re-ID) instead of person identification, that is, the query person may not occur in the gallery set. In the generalized perspective, application-driven methods that are designed for specific applications are defined as generalized open-world re-ID. Their settings are usually close to realistic application requirements. Specifically, this survey mainly includes the following four points for open-world re-ID: 1) analyzing the discrepancies between closed- and open-world scenarios; 2) describing the developments of existing open-set re-ID works and their limitations; 3) introducing specific application-driven works from three aspects, namely, raw data, practical procedure, and efficiency; and 4) summarizing the state-of-the-art methods and future directions for open-world re-ID. This survey on open-world re-ID provides a guidance for improving the usability of re-ID technique in practical applications.

Visible thermal person re-identification (VT-REID) is a task of matching person images captured by thermal and visible cameras, which is an extremely important issue in night-time surveillance applications. Existing cross-modality recognition works mainly focus on learning sharable feature representations to handle the cross-modality discrepancies. However, apart from the cross-modality discrepancy caused by different camera spectrums, VT-REID also suffers from large cross-modality and intra-modality variations caused by different camera environments and human poses, and so on. In this paper, we propose a dual-path network with a novel bi-directional dual-constrained top-ranking (BDTR) loss to learn discriminative feature representations. It is featured in two aspects: 1) end-to-end learning without extra metric learning step and 2) the dual-constraint simultaneously handles the cross-modality and intra-modality variations to ensure the feature discriminability. Meanwhile, a bi-directional center-constrained top-ranking (eBDTR) is proposed to incorporate the previous two constraints into a single formula, which preserves the properties to handle both cross-modality and intra-modality variations. The extensive experiments on two cross-modality re-ID datasets demonstrate the superiority of the proposed method compared to the state-of-the-arts.

Person re-identification (Re-ID) has been widely studied by learning a discriminative feature representation with a set of well-annotated training data. Existing models usually assume that all the training samples are correctly annotated. However, label noise is unavoidable due to false annotations in large-scale industrial applications. Different from the label noise problem in image classification with abundant samples, the person Re-ID task with label noise usually has very limited annotated samples for each identity. In this paper, we propose a robust deep model, namely PurifyNet, to address this issue. PurifyNet is featured in two aspects: 1) it jointly refines the annotated labels and optimizes the neural networks by progressively adjusting the predicted logits, which reuses the wrong labels rather than simply filtering them; 2) it can simultaneously reduce the negative impact of noisy labels and pay more attention to hard samples with correct labels by developing a hard-aware instance re-weighting strategy. With limited annotated samples for each identity, we demonstrate that hard sample mining is crucial for label corrupted Re-ID task, while it is usually ignored in existing robust deep learning methods. Extensive experiments on three datasets demonstrate the robustness of PurifyNet over the competing methods under various settings. Meanwhile, we show that it consistently improves the unsupervised/video-based Re-ID methods. Code is available at: https://github.com/mangye16/ReID-Label-Noise.

Night-time pedestrian retrieval is a cross-modality retrieval task of retrieving person images between day-time visible images and night-time thermal images. It is a very challenging problem due to modality difference, camera variations, and person variations, but it plays an important role in night-time video surveillance. The existing cross-modality retrieval usually focuses on learning modality sharable feature representations to bridge the modality gap. In this article, we propose to utilize auxiliary information to improve the retrieval performance, which consistently improves the performance with different baseline loss functions. Our auxiliary information contains two major parts: cross-modality feature distribution and contextual information. The former aligns the cross-modality feature distributions between two modalities to improve the performance, and the latter optimizes the cross-modality distance measurement with the contextual information. We also demonstrate that abundant annotated visible pedestrian images, which are easily accessible, help to improve the cross-modality pedestrian retrieval as well. The proposed method is featured in two aspects: the auxiliary information does not need additional human intervention or annotation; it learns discriminative feature representations in an end-to-end deep learning manner. Extensive experiments on two cross-modality pedestrian retrieval datasets demonstrate the superiority of the proposed method, achieving much better performance than the state-of-the-arts.

Person re-identification (Re-ID) aims at matching person images captured in non-overlapping camera views. To represent person appearance, low-level visual features are sensitive to environmental changes, while high-level semantic attributes, such as "short-hair" or "long-hair", are relatively stable. Hence, researches have started to design semantic attributes to reduce the visual ambiguity. However, to train a prediction model for semantic attributes, it requires plenty of annotations, which are hard to obtain in practical large-scale applications. To alleviate the reliance on annotation efforts, we propose to incrementally generate Deep Hidden Attribute (DHA) based on baseline deep network for newly uncovered annotations. In particular, we propose an auto-encoder model that can be plugged into any deep network to mine latent information in an unsupervised manner. To optimize the effectiveness of DHA, we reform the auto-encoder model with additional orthogonal generation module, along with identity-preserving and sparsity constraints. 1) Orthogonally generating: In order to make DHAs different from each other, Singular Vector Decomposition (SVD) is introduced to generate DHAs orthogonally. 2) Identity-preserving constraint: The generated DHAs should be distinct for telling different persons, so we associate DHAs with person identities. 3) Sparsity constraint: To enhance the discriminability of DHAs, we also introduce the sparsity constraint to restrict the number of effective DHAs for each person. Experiments conducted on public datasets have validated the effectiveness of the proposed network. On two large-scale datasets, i.e., Market-1501 and DukeMTMC-reID, the proposed method outperforms the state-of-the-art methods.

Visible thermal person re-identification (VT-ReID) is a challenging cross-modality pedestrian retrieval problem due to the large intra-class variations and modality discrepancy across different cameras. Existing VT-ReID methods mainly focus on learning cross-modality sharable feature representations by handling the modality-discrepancy in feature level. However, the modality difference in classifier level has received much less attention, resulting in limited discriminability. In this paper, we propose a novel modality-aware collaborative ensemble (MACE) learning method with middle-level sharable two-stream network (MSTN) for VT-ReID, which handles the modality-discrepancy in both feature level and classifier level. In feature level, MSTN achieves much better performance than existing methods by capturing sharable discriminative middlelevel features in convolutional layers. In classifier level, we introduce both modality-specific and modality-sharable identity classifiers for two modalities to handle the modality discrepancy. To utilize the complementary information among different classifiers, we propose an ensemble learning scheme to incorporate the modality sharable classifier and the modality specific classifiers. In addition, we introduce a collaborative learning strategy, which regularizes modality-specific identity predictions and the ensemble outputs. Extensive experiments on two cross-modality datasets demonstrate that the proposed method outperforms current state-of-the-art by a large margin, achieving rank- 1/mAP accuracy 51.64%/50.11% on the SYSU-MM01 dataset, and 72.37%/69.09% on the RegDB dataset.

Accurate prediction of mortality risk is important for evaluating early treatments, detecting high-risk patients and improving healthcare outcomes. Predicting mortality risk from the irregular clinical time series data is challenging due to the varying time intervals in the consecutive records. Existing methods usually solve this issue by generating regular time series data from the original irregular data without considering the uncertainty in the generated data, caused by varying time intervals. In this paper, we propose a novel Uncertainty-Aware Convolutional Recurrent Neural Network (UA-CRNN), which incorporates the uncertainty information in the generated data to improve the mortality risk prediction performance. To handle the complex clinical time series data with sub-series of different frequencies, we propose to incorporate the uncertainty information into the sub-series level rather than the whole time series data. Specifically, we design a novel hierarchical uncertainty-aware decomposition layer (UADL) to adaptively decompose time series into different sub-series and assign them proper weights according to their reliabilities. Experimental results on two real-world clinical datasets demonstrate that the proposed UA-CRNN method significantly outperforms state-of-the-art methods in both short-term and long-term mortality risk predictions.

This paper studies the unsupervised embedding learning problem, which requires an effective similarity measurement between samples in low-dimensional embedding space. Motivated by the positive concentrated and negative separated properties observed from category-wise supervised learning, we propose to utilize the instance-wise supervision to approximate these properties, which aims at learning data augmentation invariant and instance spread-out features. To achieve this goal, we propose a novel instance based softmax embedding method, which directly optimizes the `real' instance features on top of the softmax function. It achieves significantly faster learning speed and higher accuracy than all existing methods. The proposed method performs well for both seen and unseen testing categories with cosine similarity. It also achieves competitive performance even without pre-trained network over samples from fine-grained categories.

This paper addresses the variation generalized feature learning problem in unsupervised video-based person re-identification (re-ID). With advanced tracking and detection algorithms, large-scale intra-view positive samples can be easily collected by assuming that the image frames within the tracking sequence belong to the same person. Existing methods either directly use the intra-view positives to model cross-view variations or simply minimize the intra-view variations to capture the invariant component with some discriminative information loss. In this paper, we propose a Variation Generalized Feature Learning (VGFL) method to learn adaptable feature representation with intra-view positives. The proposed method can learn a discriminative re-ID model without any manually annotated cross-view positive sample pairs. It could address the unseen testing variations with a novel variation generalized feature learning algorithm. In addition, an Adaptability-Discriminability (AD) fusion method is introduced to learn adaptable video-level features. Extensive experiments on different datasets demonstrate the effectiveness of the proposed method.

Visible thermal person re-identification (VT-ReID) is a cross-modality pedestrian retrieval problem, which automatically searches persons between day-time visible images and night-time thermal images. Despite the extensive progress in single-modality ReID, the cross-modality pedestrian retrieval problem has limited attention due to its challenges in modality discrepancy and large intra-class variations across cameras. Existing cross-modality ReID methods usually solve this problem by learning cross-modality feature representations with modality-sharable classifier. However, this learning strategy may lose discriminative information in different modalities. In this paper, we propose a novel modality-aware collaborative (MAC) learning method on top of a two-stream network for VT-ReID, which handles the modality-discrepancy in both feature level and classifier level. In feature level, it handles the modality discrepancy by a two-stream network with different parameters. In classifier level, it contains two separate modality-specific identity classifiers for two modalities to capture the modality-specific information, and they have the same network architecture but different parameters. In addition, we introduce a collaborative learning scheme, which regularizes the modality-sharable and modality-specific identity classifiers by utilizing the relationship between different classifiers. Extensive experiments on two cross-modality person re-identification datasets demonstrate the superiority of the proposed method, achieving much better performance than the state-of-the-art.

Cross-camera label estimation from a set of unlabelled training data is an extremely important component in unsupervised person re-identification (re-ID) systems. With the estimated labels, existing advanced supervised learning methods can be leveraged to learn discriminative re-ID models. In this paper, we utilize the graph matching technique for accurate label estimation due to its advantages in optimal global matching and intra-camera relationship mining. However, the graph structure constructed with non-learnt similarity measurement cannot handle the large cross-camera variations, which leads to noisy and inaccurate label outputs. This paper designs a Dynamic Graph Matching (DGM) framework, which improves the label estimation process by iteratively refining the graph structure with better similarity measurement learnt from intermediate estimated labels. In addition, we design a positive re-weighting strategy to refine the intermediate labels, which enhances the robustness against inaccurate matching output and noisy initial training data. To fully utilize the abundant video information and reduce false matchings, a co-matching strategy is further incorporated into the framework. Comprehensive experiments conducted on three video benchmarks demonstrate that DGM outperforms state-of-the-art unsupervised re-ID methods and yields competitive performance to fully supervised upper bounds.

Tracking target of interests is an important step for motion perception in intelligent video surveillance systems. While most recently developed tracking algorithms are grounded in RGB image sequences, it should be noted that information from RGB modality is not always reliable (e.g. in a dark environment with poor lighting condition), which urges the need to integrate information from infrared modality for effective tracking because of the insensitivity to illumination condition of infrared thermal camera. However, several issues encountered during the tracking process limit the fusing performance of these heterogeneous modalities: 1) the cross-modality discrepancy of visual and motion characteristics, 2) the uncertainty of degree of reliability in different modalities, and 3) large target appearance variations and background distractions within each modality. To address these issues, this paper proposes a novel and optimal discriminative learning framework for multi-modality tracking. In particular, the proposed discriminative learning framework is able to: 1) jointly eliminate outlier samples caused by large variations and learn discriminability-consistent features from heterogeneous modalities, and 2) collaboratively perform modality reliability measurement and target-background separation. Extensive experiments on RGB-infrared image sequences demonstrate the effectiveness of the proposed method.

Person re-identification is widely studied in visible spectrum, where all the person images are captured by visible cameras. However, visible cameras may not capture valid appearance information under poor illumination conditions, e.g, at night. In this case, thermal camera is superior since it is less dependent on the lighting by using infrared light to capture the human body. To this end, this paper investigates a cross-modal re-identification problem, namely visible-thermal person re-identification (VT-REID). Existing cross-modal matching methods mainly focus on modeling the cross-modality discrepancy, while VT-REID also suffers from cross-view variations caused by different camera views. Therefore, we propose a hierarchical cross-modality matching model by jointly optimizing the modality-specific and modality-shared metrics. The modality-specific metrics transform two heterogenous modalities into a consistent space that modality-shared metric can be subsequently learnt. Meanwhile, the modality-specific metric compacts features of the same person within each modality to handle the large intra-modality intra-person variations (e.g. viewpoints, pose). Additionally, an improved two-stream CNN network is presented to learn the multi-modality sharable feature representations. Identity loss and contrastive loss are integrated to enhance the discriminability and modality-invariance with partially shared layer parameters. Extensive experiments illustrate the effectiveness and robustness of the proposed method.

Person re-identification (REID) is an important task in video surveillance and forensics applications. Most of previous approaches are based on a key assumption that all person images have uniform and sufficiently high resolutions. Actually, various low-resolutions and scale mismatching always exist in open world REID. We name this kind of problem as Scale-Adaptive Low Resolution Person Re-identification (SALR-REID). The most intuitive way to address this problem is to increase various low-resolutions (not only low, but also with different scales) to a uniform high-resolution. SR-GAN is one of the most competitive image super-resolution deep networks, designed with a fixed upscaling factor. However, it is still not suitable for SALR-REID task, which requires a network not only synthesizing high-resolution images with different upscaling factors, but also extracting discriminative image feature for judging person’s identity. (1) To promote the ability of scale-adaptive upscaling, we cascade multiple SRGANs in series. (2) To supplement the ability of image feature representation, we plug-in a reidentification network. With a unified formulation, a Cascaded Super-Resolution GAN (CSR-GAN) framework is proposed. Extensive evaluations on two simulated datasets and one public dataset demonstrate the advantages of our method over related state-of-the-art methods.

Cross-modality person re-identification between the thermal and visible domains is extremely important for night-time surveillance applications. Existing works in this filed mainly focus on learning sharable feature representations to handle the cross-modality discrepancies. However, besides the cross-modality discrepancy caused by different camera spectrums, visible thermal person re-identification also suffers from large cross-modality and intra-modality variations caused by different camera views and human poses. In this paper, we propose a dual-path network with a novel bi-directional dual-constrained top-ranking loss to learn discriminative feature representations. It is advantageous in two aspects: 1) end-to-end feature learning directly from the data without extra metric learning steps, 2) it simultaneously handles the cross-modality and intra-modality variations to ensure the discriminability of the learnt representations. Meanwhile, identity loss is further incorporated to model the identity-specific information to handle large intra-class variations. Extensive experiments on two datasets demonstrate the superior performance compared to the state-of-the-arts.

Person re-identifcation is a key technique to match person images captured in non-overlapping camera views. Due to the sensitivity of visual features to environmental changes, semantic attributes, such as "short-hair" or "long-hair", begin to be investigated to represent person's appearance to improve the re-identifcation performance. Generally, training semantic attribute representations requires massive annotated samples, which limits the applicability on the large-scale practical applications. To alleviate the reliance on annotation efforts, we propose a new person representation with hidden attributes by mining latent information from visual feature in an unsupervised manner. In particular, an auto-encoder model is plugged-in to the deep learning network to compose a Deep Hidden Attribute Network (DHA-Net). The learnt hidden attribute representation preserves the robustness of semantic attributes and simultaneously inherits the discrimination ability of visual features. Experiments conducted on public datasets have validated the effectiveness of DHA-Net. On two large-scale datasets, i.e., Market-1501 and DukeMTMC-reID, the proposed method outperforms the state-of-the-art methods.

Label estimation is an important component in an unsupervised person re-identification (re-ID) system. This paper focuses on cross-camera label estimation, which can be subsequently used in feature learning to learn robust re-ID models. Specifically, we propose to construct a graph for samples in each camera, and then graph matching scheme is introduced for cross-camera labeling association. While labels directly output from existing graph matching methods may be noisy and inaccurate due to significant cross-camera variations, this paper propose a dynamic graph matching (DGM) method. DGM iteratively updates the image graph and the label estimation process by learning a better feature space with intermediate estimated labels. DGM is advantageous in two aspects: 1) the accuracy of estimated labels is improved significantly with the iterations; 2) DGM is robust to noisy initial training data. Extensive experiments conducted on three benchmarks including the large-scale MARS dataset show that DGM yields competitive performance to fully supervised baselines, and outperforms competing unsupervised learning methods.1

Saliency detection for images and videos becomes increasingly popular due to its wide applicability. Enormous research efforts have been focused on saliency detection, but it still has some issues in maintaining spatiotemporal consistency of videos and uniformly highlighting entire objects. To address these issues, this paper proposes a superpixel-level spatiotemporal saliency model for saliency detection in videos. To detect salient object, we extract multiple spatiotemporal features combined with intra-consistency motion information preliminarily. Meanwhile, considering inter-consistency of foreground in videos, a set of foreground locations are obtained from previous frames. Then, we introduce foreground-background and local foreground contrast saliency cues of those features using the location prior information of foreground. These two improved contrast saliency cues uniformly highlight the entire object and suppress the background effectively. Finally, we use an interactively dynamic fusion method to integrate the output spatial and temporal saliency maps. The proposed approach is validated on challenging sets of video sequences. Subjective observations and objective evaluations demonstrate that the proposed model achieves a better performance on saliency detection compared with the state-of-the-art spatiotemporal saliency methods.

Person re-identification, aiming to identify images of the same person from various cameras configured in different places, has attracted much attention in the multimedia retrieval community. In this problem, choosing a proper distance metric is a crucial aspect, and many classic methods utilize a uniform learnt metric. However, their performance is limited due to ignoring the zero-shot and fine-grained characteristics presented in real person re-identification applications. In this paper, we investigate two consistencies across two cameras, which are cross-view support consistency and cross-view projection consistency. The philosophy behind it is that, in spite of visual changes in two images of the same person under two camera views, the support sets in their respective views are highly consistent, and after being projected to the same view, their context sets are also highly consistent. Based on the above phenomena, we propose a data-driven distance metric (DDDM) method, re-exploiting the training data to adjust the metric for each query-gallery pair. Experiments conducted on three public data sets have validated the effectiveness of the proposed method, with a significant improvement over three baseline metric learning methods. In particular, on the public VIPeR dataset, the proposed method achieves an accuracy rate of 42.09% at rank-1, which outperforms the state-of-the-art methods by 4.29%.

Person reidentification is a key technique to match different persons observed in nonoverlapping camera views. Many researchers treat it as a special object-retrieval problem, where ranking optimization plays an important role. Existing ranking optimization methods mainly utilize the similarity relationship between the probe and gallery images to optimize the original ranking list, but seldom consider the important dissimilarity relationship. In this paper, we propose to use both similarity and dissimilarity cues in a ranking optimization framework for person reidentification. Its core idea is that the true match should not only be similar to those strongly similar galleries of the probe, but also be dissimilar to those strongly dissimilar galleries of the probe. Furthermore, motivated by the philosophy of multiview verification, a ranking aggregation algorithm is proposed to enhance the detection of similarity and dissimilarity based on the following assumption: the true match should be similar to the probe in different baseline methods. In other words, if a gallery blue image is strongly similar to the probe in one method, while simultaneously strongly dissimilar to the probe in another method, it will probably be a wrong match of the probe. Extensive experiments conducted on public benchmark datasets and comparisons with different baseline methods have shown the great superiority of the proposed ranking optimization method.

Searching for specific persons from surveillance videos captured by different cameras, is a key yet under-addressed challenge in multimedia system. Related person retrieval works mainly focus on searching person by visual appearance, known as person re-identification. However, the initial visual image may not be available in some practical applications. For example, the criminal is described by a text description indirectly, "A young woman wearing a red casual with a backpack", the traditional methods can not conquer this issue. Based on a set of pre-defined attributes that the text description query can be transformed to an attribute vector, thus can be used to retrieval in the gallery set. And yet, the user-provided attributes are sometimes incomplete. This new issue is defined as Specific Person Retrieval via Incomplete Text Description. In this paper, we conduct a specific attribute completion to enrich the original text query and generate a more expressive attribute vector. Then, a pairwise-based metric learning is introduced for completed attribute vectors. Extensive experiments conducted on two benchmark datasets have shown our superior performance.

Person re-identification is a key technique to match different persons observed in non-overlapping camera views.Many researchers treat it as a special object retrieval problem, where ranking optimization plays an important role. Existing ranking optimization methods utilize the similarity relationship between the probe and gallery images to optimize the original ranking list in which dissimilarity relationship is seldomly investigated. In this paper, we propose to use both similarity and dissimilarity cues in a ranking optimization framework for person re-identification. Its core idea is based on the phenomenon that the true match should not only be similar to the strong similar samples of the probe but also dissimilar to the strong dissimilar samples. Extensive experiments have shown the great superiority of the proposed ranking optimization method.