Large language models (LLMs) are foundational to artificial general intelligence (AGI), while imposing stringent hardware demands in computational power and memory bandwidth. To meet these demands, recent advances in hybrid bonding offer new opportunities through high-bandwidth, low-latency logic-memory 3D integration. Due to the spatial distribution of bandwidth in 3D-stacked DRAM, accesses to remote DRAM banks need to traverse the network-on-chip (NoC). Since GEMM is the dominant operator in LLM workloads, we analyze existing distributed GEMM algorithms and show that they suffer from communication inefficiency on mesh topologies. To address these challenges, we present FlashGEMM, a mesh-aware distributed GEMM algorithm that achieves efficient communication-computation overlap and optimized execution flow for 3D-stacked LLM accelerators. Evaluations show that FlashGEMM delivers up to 1.50× improvement in time-to-first-token (TTFT) and 7.11× improvement in throughput over prior methods.

We address event-based deblurring for low-light imaging, where conventional frames suffer severe blur, noise and saturation, while events capture sharp high-frequency contrast changes with microsecond latency that can guide the recovery of lost structures. Existing event-based reconstruction methods neither explicitly model low-light noise and saturation nor enforce precise alignment between events and frames, which limits cross-modal fusion and deblurring quality. We propose the Wiener-Deconvolution-Driven Event-Based Deblurring Network (WiED-Net), which embeds the Wiener deconvolution into a deep architecture so that the physical imaging model and noise statistics are encoded in the frequency domain and high-frequency recovery is stabilized on noise dominated night data. WiED-Net adopts a two stage design. The first stage applies Wiener deconvolution in both image and feature spaces to suppress noise, recover saturated regions and reduce ringing, assisted by an eventguided cross-modal feature fusion (ECFF) module for accurate alignment. The second stage uses a multi-scale fusion module to integrate the complementary event and image branches. Training is constrained by a set of losses, including a tailored blur kernel loss that provides closed-loop regularization from physical priors. Together, these designs enable WiED-Net to recover fine details while robustly suppressing artifacts and noise, and to achieve superior quantitative and qualitative performance, achieving superior quantitative and qualitative performance with a notable improvement of 1.97 dB in PSNR and 5% in SSIM over the previous state-of-the-art methods in low-light deblurring. Code will be available at https://github.com/zhuzifeng38/WiED-Net.

Tokenized pose estimation (TPE) has demonstrated remarkable performance in lightweight human pose estimation (HPE) models. However, existing TPE methods typically initialize keypoint tokens randomly, without explicitly incorporating human structure priors. These priors play a vital role in HPE by effectively mitigating common challenges such as occlusion and ambiguity. To this end, we propose a Structure-Aware Keypoint Position Embedding (SAKPE). This embedding explicitly encodes inherent structural properties of the human body, such as symmetry and order, into the positional coordinates of keypoint tokens. It also employs learnable scale and offset factors to adapt to diverse human poses, thereby fully exploiting the geometric constraints among keypoints. Furthermore, to better leverage the positional relationships among patch tokens, we introduce a Layer-adaptive Hybrid Patch Position Embedding (LHPPE). It dynamically fuses absolute and relative position embeddings of patch tokens based on attention distributions across Transformer layers, enabling the model to learn both absolute and relative positional information adaptively. Taking the two together, we propose a novel position embedding method for pose estimation, named Human-structure-aware Token Position Embedding (HTPE). It significantly improves the performance of various TPE models. Extensive experiments on COCO, CrowdPose, and OCHuman show that HTPE achieves state-of-the-art (SOTA) performance among lightweight methods, with a negligible increase in parameters and FLOPs. Notably, it demonstrates consistent improvements under occlusion,, achieving up to 3.3 AP gains. The source code can be found in https://github.com/guzejungithub/HTPE.

Existing image restoration and enhancement (IRE) methods suffer from three fundamental limitations: 1) they present a high technical barrier, requiring expert knowledge and lacking intuitive natural language control; 2) they are inflexible and poorly adaptable, as models are typically designed for single, specific degradations and fail on complex or mixed real-world scenarios; and 3) they lack interactivity and ignore subjectivity, operating as "closed-box" tools that cannot incorporate human feedback or understand nuanced user intentions. To overcome these challenges, we pioneer a novel paradigm: a Multi-Agent System (MAS) for interactive and adaptive image restoration. We design and implement a prototype system, Interactive and Adaptive Multi-Agent System (IAMAgent), which orchestrates a team of specialized agents to collaboratively solve complex IRE tasks. At its core, a Manager Agent, driven by a Large Language Model, interprets user commands, devises strategies, and allocates sub-tasks. It directs a Perception Agent for degradation diagnosis, a suite of specialized Execution Agents that encapsulate various low-level vision models, and a Critique Agent for automated quality assessment. This collaborative framework enables an innovative, language-driven, and human-in-the-loop optimization process. Our work is the first to introduce the MAS paradigm to the IRE domain, transforming it from a collection of static tools into a dynamic, user-centric, and intelligent system. We demonstrate that IAMAgent not only significantly enhances restoration performance and adaptability but also bridges the critical gap between high-level human intention and low-level vision tasks.

In practical applications of human pose estimation, low-resolution inputs frequently occur, and existing state-of-the-art models perform poorly with low-resolution images. This work focuses on boosting the performance of low-resolution models by distilling knowledge from a high-resolution model. However, we face the challenge of feature size mismatch and class number mismatch when applying knowledge distillation to networks with different input resolutions. To address this issue, we propose a novel cross-domain knowledge distillation (CDKD) framework. In this framework, we construct a scale-adaptive projector ensemble (SAPE) module to spatially align feature maps between models of varying input resolutions. It adopts a projector ensemble to map low-resolution features into multiple common spaces and adaptively merges them based on multi-scale information to match high-resolution features. Additionally, we construct a cross-class alignment (CCA) module to solve the problem of the mismatch of class numbers. By combining an easy-to-hard training (ETHT) strategy, the CCA module further enhances the distillation performance. The effectiveness and efficiency of our approach are demonstrated by extensive experiments on three common benchmark datasets: MPII, COCO, and Crowdpose.

Transformer-based Single Image Deraining (SID) methods have achieved remarkable success, primarily attributed to their robust capability in capturing long-range interactions. However, we've noticed that current methods handle rain-affected and unaffected regions concurrently, overlooking the disparities between these areas, resulting in confusion between rain streaks and background parts, and inabilities to obtain effective interactions, ultimately resulting in suboptimal deraining outcomes. To address the above issue, we introduce the Region Transformer (Regformer), a novel SID method that underlines the importance of independently processing rain-affected and unaffected regions while considering their combined impact for high-quality image reconstruction. The crux of our method is the innovative Region Transformer Block (RTB), which integrates a Region Masked Attention (RMA) mechanism and a Mixed Gate Forward Block (MGFB). Our RTB is used for attention selection of rain-affected and unaffected regions and local modeling of mixed scales. The RMA generates attention maps tailored to these two regions and their interactions, enabling our model to capture comprehensive features essential for rain removal. To better recover high-frequency textures and capture more local details, we develop the MGFB as a compensation module to complete local mixed scale modeling. Extensive experiments demonstrate that our model reaches state-of-the-art performance, significantly improving the image deraining quality. Our code and trained models are publicly available athttps://github.com/ztMotaLee/Regformer.

Event cameras hold great potential for motion deblurring because they capture motion information with microsecond precision, offering robustness to motion blur. However, the limited interaction between RGB frames and event streams presents a significant challenge, preventing the full utilization of the event cameras' unique advantages. To address this, we proposeDual frame-eventInteraction and introduce a multi-scaleNetwork structure, DuInt-Net. DuInt-Net aims to tackle two key challenges: (1) enhancing the representational and interaction capabilities between RGB frames and event streams, and (2) adaptively selecting richer visual features for improved motion deblurring. We introduce an event-frame joint interaction module that consists of three branches: a base branch, a global awareness attention branch, and a local enhancement attention branch. The base branch processes essential pixel-level features that retain the original structural information. The global branch integrates event data to improve large-scale motion understanding, while the local branch uses large-kernel convolutions to refine fine-grained details in RGB frames. For superior reconstruction performance, we also propose the event-guided multi-scale fusion attention module, which effectively combines local visual information and global frame-event relationships. Extensive experiments demonstrate that DuInt-Net achieves superior performance, both quantitatively and qualitatively, showcasing its superior motion deblurring capabilities.

Recently, there has been a surge in the development of robust norm distance-based linear discriminant analysis (LDA) techniques, which have garnered significant attention in the field of feature extraction. However, a persistent issue that has yet to be resolved is that the successful suppression of outliers may inadvertently impede the accurate discrimination of normal points. To solve this problem, we, in this article, study a novel robust LDA measured by double capped $L_{p}$ -norm distance (CLD) metrics with min constraints (DCLDA) to learn robust discriminant projections, in which normal points and outliers are separately treated. To be specific, it takes a double capped $L_{p}$ -norm with "Min" constraints in the proposed model to measure the distances for between- and within-class dispersions. The proposed model effectively ensures accurate discrimination of normal points by $L_{p}$ -norm, while also eliminating the exaggerated effect of outliers that may arise from larger $p$ values. The resulted objective is not trivial because of its nonconvexity and nonsmoothness. As one of the major contributions of this article, we introduce a new reformulation that provides an objective problem theoretically equivalent to the original. By this reformulation, we develop an effective iterative algorithm to solve the proposed model. The algorithm is proven to be convergent through rigorous theoretical analysis. Extensive experiments were conducted on several real-world datasets across different image classification tasks to showcase the effectiveness of the proposed method.

Offline signature verification remains vulnerable to real-world degradations such as low-resolution scans, transmission artifacts, and environmental noise. We introduce a variant robustness mechanism that foregrounds stroke-level evidence, and instantiate it in a Diffusion-based Robust Signature Verification (DRSV) framework. DRSV integrates two components: (i) a diffusion-driven noise-signature relation module that synthesizes plausible variations to enhance structural resilience, and (ii) a transformer-based verifier that fuses the predicted noise with the signature image via Bi-Directional Channel-Wise Correlation Attention (BCCA) to decide authenticity. Experiments on five public datasets (CEDAR, BHSig–B, BHSig–H, GPDS Synthetic, UTSig) show consistent accuracy gains and improved robustness across diverse perturbations, indicating the effectiveness of the proposed framework.

The garment structure serves as a crucial medium for expressing the designer's creative vision and showcasing the distinctive character of clothing items. Effective editing of garment structure in fashion images allows for an advanced preview of the design, accelerating the process of garment customization to meet individualized requirements. Although large-scale diffusion models have demonstrated impressive image generation and editing capabilities, no efforts have been made to exploit their potential in part-level editing of images. Unlike previous research, we define a clothing structure editing (CSE) task aimed at accurately editing the local structure of human-centered clothing images through simple instruction-based prompts while maintaining the consistency of clothing appearance. Specifically, this paper develops a new controllable triple-flow framework for structure editing named FashionTailor. An additional network called ClothingNet is proposed to extract the clothing details to address the rigid constraints of the original garment structure. Then, we propose a semantic-refined module to extract the semantic understanding of the source image and adaptively focus on the part to be edited. We also design a cross-blend attention mechanism to integrate fine-grained clothing features to guarantee precise alignment between appearance and target structure features. In addition, a garment structure dataset called StructureFashion has been collated, wherein each item of clothing is represented by multiple photos with diverse structure characteristics, containing over six million pairs. Finally, our method supports editing the structure of multiple parts on a garment simultaneously. Extensive experiments validate the effectiveness of our method for editing part-level human images in StructureFashion dataset and real-scenarios.

Existing face forgery detection methods achieve promising performance when training and testing forgery data are from identical manipulation types, while they fail to generalize well to unseen samples. In this paper, we experimentally investigate and find that the poor generalization of the methods mainly arises from their overfitting on the known fake patterns. Excessively focused on seen fakes, those detectors fail to effectively learn image-intrinsic information and the distributional disparity between real and fake images. Then, to address this issue, we redefine fake learning as real-fake distributional disparity learning. We propose a novel deepfake detection framework learning distributional disparity based on the differentiated reconstruction on real and fake images for improved generalization. Specifically, distributional disparity learning on differentiated reconstruction of the real and fake images, enforces the model to learn image-invariant intrinsic representations. The reconstruction on real and fake images forces the decoders to learn the distribution of real and fake images, respectively. Moreover, to avoid the influence from the specificalization of the known fake patterns, we further propose the information interaction learning on the encoded intrinsic information and the pixel disparity between the input image and its reconstruction to distinguish face forgeries that are even unknown. Extensive experiments on large-scale benchmark datasets demonstrated the effectiveness of addressing the overfitting issue of the classification network, and verified the superior performance of our method.

This paper tackles the challenge of stereoscopic image rain removal by focusing on enhancing texture integrity and disparity consistency. Existing stereoscopic rain removal techniques often fall short due to 1) disruptions in texture coherence caused by complex rain streaks, and 2) inaccuracies in disparity estimation from inadequate feature fusion. To overcome these limitations, we introduce the StereoIRR method, which incorporates: 1) a Long-range and Cross-view Interaction (LCI) framework that preserves texture integrity by mitigating rain’s adverse effects on stereoscopic features, and 2) a Dual-view Mutual Attention mechanism that ensures disparity consistency by generating precise mutual attention maps for cross-view feature fusion. Our approach not only maintains the integrity of stereoscopic textures but also significantly reduces errors in disparity estimation. Extensive experiments demonstrate that StereoIRR consistently outperforms state-of-the-art monocular and stereoscopic methods on multiple benchmark datasets.

Many consumer cameras are equipped with 8-bit image sensors, which often struggle to capture scenes with a High Dynamic Range (HDR). This limitation can result in overexposed or underexposed regions, a loss of fine details due to low bit-depth compression, skewed color distributions, and noticeable noise in dark areas. Traditional Standard Dynamic Range (SDR) image enhancement methods typically focus on color mapping by expanding the color range and adjusting brightness. However, they often fail to restore details in dynamic range extremes, i.e. regions where pixel values approach the minimum or maximum limits. We define “exposure-limited image enhancement” as the process of enhancing images with large missing areas due to exposure issues within the SDR space, which differs from existing “mis-exposed image enhancement” methods primarily aimed at correcting color distributions. To enhance these exposure-limited images and overcome the limitations of current models, we propose a novel two-stage approach. In the first stage, we remap color and brightness to a suitable range while preserving existing details. In the second stage, we use a diffusion prior to generate content in severely overexposed or underexposed regions, which are otherwise lost during capture. Notably, this generative refinement module can also serve as a plug-and-play component alongside existing enhancement methods. Extensive experiments demonstrate that our method significantly improves image quality and detail, outperforming state-of-the-art techniques in dynamic range extremes. The project page is at https://Sagiri0208.github.io.

Image segmentation has found widespread applications in computer vision, particularly in fields such as medical image analysis, autonomous driving, and video surveillance. However, as the complexity of segmentation tasks increases, new vision foundation models have emerged, notably the Segment Anything Model (SAM) series. By leveraging large-scale training and cross-domain generalization, SAM2 has significantly enhanced the performance of image segmentation. However, despite these advancements, SAM2 still generates class-agnostic segmentation results in the absence of manual prompts. To enhance the SAM2's performance in these scenarios, we propose an effective framework, termed SAM2-Cap. Specifically, we integrate the backbone of SAM2 with a Capsule Autoencoder, exploiting the inherent properties of capsule networks to effectively capture and decode the spatial hierarchical relationships between the parts and objects. This improves the modeling of object shapes and spatial relationships, enhancing the understanding and segmentation accuracy of complex objects while reducing the reliance on prompt information. Extensive experiments on Prostate MRI segmentation, Polyp segmentation and Cityscapes dataset, demonstrate our method achieves competitive results compared to state-of-the-art models.

Estimating 3D human poses from monocular videos is a challenging task, primarily due to self-occlusion. Many existing methods struggle with unseen viewpoints as they rely on large amounts of data rather than enhancing their generalization ability across different viewpoints. To overcome this limitation, we propose a novel approach using a capsule autoencoder integrated with the shifted-windows model (SwinCAE), which can enhance prediction accuracy by effectively capturing the spatial hierarchical relationship between the parts and objects. Furthermore, we build a Parallel Double Attention with Shifted Windows module to enhance computational efficiency and modeling capacity. Additionally, we construct a Multi-Attention Collaborative module to capture diverse information, including both coarse and fine details. Through the collaboration of these modules, the model representation is significantly improved, resulting in a more accurate generated pose. Extensive experiments demonstrate that SwinCAE achieves better or comparable results to state-of-the-art models about 3D human pose estimation task.

Recently, 3D Gaussian Splatting (3DGS) has achieved remarkable results in 3D reconstruction and view synthesis tasks. However, single-view feed-forward 3DGS still faces significant challenges. Current state-of-the-art (SOTA) single-view 3DGS methods typically employ a small number of layers (1-2 layers) with Gaussian Splatting (GS) representations at the same resolution as the input image to address the irregularity of GS data. However, such shallow and uniform GS primitive distributions is difficult to represent occluded regions and important spatial details. Inspired by multi-plane images, this paper proposes a Multi-Layer Gaussian Splatting (MLGS) representation, which consists of shallow base GS layers for visible content and multiple occlusion GS layers dedicated to reconstructing occluded regions. The proposed MLGS representation explicitly decouples the learning processes of visible and occluded content while enhancing occlusion prediction through the following components. First, spatial stratification of GS is achieved by estimating the depth distribution range of GS primitives across different layers, forcing GS to learn spatial content reconstruction at different depths. Second, a mask-guided mechanism is proposed to effectively isolate occlusion regions and guide inpainting using spatially context-aware features. Finally, a gated convolution block is designed to dynamically modulate feature fusion to enhance reconstruction fidelity. With separate loss supervision for base and occlusion layers, MLGS enables geometrically plausible scene completion. Experiments on RealEstate10K, KITTI, and NYUv2 datasets demonstrate that the proposed method achieves SOTA performance for single-image spatial scene reconstruction.

As a form of multimedia creation, visual novel (VN) conveys engaging narratives through the integrated presentation of text, images, and music, and has shown promise across various application domains. Recent advances in generative AI have fueled interest in automating VN creation using LLMs and other foundation models. However, fully end-to-end VN creation (i.e., from user description to executable VN) remains underexplored and presents several key challenges: 1) the hallucination and limited capacity of LLMs hinder the generation of long and coherent plots; 2) current models lack effective mechanisms for ensuring cross-modal consistency between plot, visual, and audio elements. To address these issues, we propose a hierarchical end-to-end framework for automatic VN generation and assembly, which employs an outline-guided autoregressive generation mechanism that transforms high-level user prompts into coherent plots, while a vision LLM-based self-correction mechanism ensures consistency between multimedia assets and plot content. Additionally, we introduce a script validation mechanism to ensure the executable of the final VN application. Experiments demonstrate that our framework generates high-quality VN applications with coherent storylines and consistent multimedia content.

Deep learning-based low-light image enhancement (LLIE) is a task of leveraging deep neural networks to enhance the image illumination while keeping the image content unchanged. From the perspective of training data, existing methods complete the LLIE task driven by one of the following three data types: paired data, unpaired data and zero-reference data. Each type of these data-driven methods has its own advantages, e.g., zero-reference data-based methods have very low requirements on training data and can meet the human needs in many scenarios. In this paper, we leverage pure Gaussian noise to complete the LLIE task, which further reduces the requirements for training data in LLIE tasks and can be used as another alternative in practical use. Specifically, we propose Noise SElf-Regression (NoiSER) without access to any task-related data, simply learns a convolutional neural network equipped with an instance-normalization layer by taking a random noise image, for each pixel, as both input and output for each training pair, and then the low-light image is fed to the trained network for predicting the normal-light image. Technically, an intuitive explanation for its effectiveness is as follows: 1) the self-regression reconstructs the contrast between adjacent pixels of the input image, 2) the instance-normalization layer may naturally remediate the overall magnitude/lighting of the input image, and 3) the assumption for each pixel enforces the output image to follow the well-known gray-world hypothesis (Buchsbaum, 1980) when the image size is big enough. Compared to current state-of-the-art LLIE methods with access to different task-related data, NoiSER is highly competitive in enhancement quality, yet with a much smaller model size, and much lower training and inference cost. In addition, the experiments also demonstrate that NoiSER has great potential in overexposure suppression and joint processing with other restoration tasks.

Recently, interest in flat-type projection clustering methods has grown as they improve learner's performance by exploring multiple projection subspaces. However, solvers used in previous representative works predominantly rely on greedy search strategies, which incur high computational costs and fail to consider interdependencies between projections. Moreover, these methods do not simultaneously guarantee the effective suppression of outliers and noisy data at cluster boundaries, ultimately compromising data discrimination. To address these limitations and discover a more effective subspace for each flat, we propose robust multiple flat projections clustering (RMFPC). This method computes within- and between-cluster distances using the L2,1-norm to enhance robustness against outliers. Furthermore, we propose a truncated distance maximization constraint (TDMC) to eliminate the influence of noisy data on cluster separability. The resulting objective is presented in a ratio form, which is not trivial. We provide a novel formulation to achieve a theoretically equivalent problem. Based on this reformulation, we develop an efficient non-greedy solution algorithm. In addition, a cluster center optimization mechanism is incorporated into the solution process to accurately estimate the distribution of each cluster center. The convergence analysis and proof of the proposed algorithm are provided. Experiments on both toy and real-world datasets demonstrate the effectiveness of the proposed method.

The evolution of individuals' living standards has transformed clothing preferences, elevating fashion beyond mere utility to a potent means of self-expression. However, the intricate task of outfit selection persists as a challenge, marked by traditional methods facing challenges such as the oversight of combined factors of scene and body shape, insufficient emphasis on detail-oriented matching, and overreliance on rigid hierarchical structures. To tackle these challenges, this article introduces a novel model, termedGlobal-Local matching network towardsOutfit recommendation for diverse bodyShapes andScenes (GLOSS). Specifically, we first introduce a newly compiled fashion dataset, StreetFashion, to capture the combined factors of body shapes and scene characteristics. Additionally, we develop innovative multi-level globality- and locality-aware matching methods to enhance the accuracy of outfit recommendations by comprehensively considering both global and local relationships among clothing items, outfits, users, and scenes. Furthermore, we develop a personalized outfit heterogeneous graph that incorporates historical interactions among fashion entities, enabling effective modeling of nonstrict hierarchical relationships. Evaluation conducted on both our collected dataset and an adapted existing dataset demonstrates the effectiveness of our proposed approach in outfit recommendation. Our codes are released at:https://github.com/ChenrMa/GLOSS.

Recent mainstream image captioning methods usually adopt two-stage captioners, i.e., calculating the object features of the given image by a pre-trained detector and then feeding them into a language model to generate the descriptive sentences. However, such a two-stage procedure will lead to a task-based information gap that decreases the performance of the captioners, because the object features learned from the detection task are suboptimal representations and cannot provide all the necessary information for subsequent sentence generation. Besides, the object features are usually represented by the last pooling features of the detector that lose the local details of images. In this paper, we propose a novel One-Stage Image Captioner using dynamic multi-sight embedding and alignment, called SeaCap, which directly transforms input images into descriptive sentences in one stage to eliminate the information gap. Specifically, to obtain rich features, we use the Swin Transformer to capture multi-level features, followed by a sights alignment module to alleviate the vision confusion, and then feed them into a novel dynamic multi-sight embedding module to exploit both the global structure and local texture of input images. To enhance the global modeling capacity of the visual encoder, we propose a new dual-dimensional refining module to non-locally model the interaction of the embedded features. As a result, SeaCap can obtain rich and useful information to improve the performance of the captioner. Extensive comparisons on the benchmark MS-COCO, Flickr8K and Flickr30 K datasets verified the superior performance of our method.

Weakly supervised temporal action localization (TAL) aims to localize the action instances in untrimmed videos using only video-level action labels. Without snippet-level labels, this task should be hard to distinguish all snippets with accurate action/background categories. The main difficulties are the large variations brought by the unconstraint background snippets and multiple subactions in action snippets. The existing prototype model focuses on describing snippets by covering them with clusters (defined as prototypes). In this work, we argue that the clustered prototype covering snippets with simple variations still suffers from the misclassification of the snippets with large variations. We propose an ensemble prototype network (EPNet), which ensembles prototypes learned with consensus-aware clustering. The network stacks a consensus prototype learning (CPL) module and an ensemble snippet weight learning (ESWL) module as one stage and extends one stage to multiple stages in an ensemble learning way. The CPL module learns the consensus matrix by estimating the similarity of clustering labels between two successive clustering generations. The consensus matrix optimizes the clustering to learn consensus prototypes, which can predict the snippets with consensus labels. The ESWL module estimates the weights of the misclassified snippets using the snippet-level loss. The weights update the posterior probabilities of the snippets in the clustering to learn prototypes in the next stage. We use multiple stages to learn multiple prototypes, which can cover the snippets with large variations for accurate snippet classification. Extensive experiments show that our method achieves the state-of-the-art weakly supervised TAL methods on two benchmark datasets, that is, THUMOS'14, ActivityNet v1.2, and ActivityNet v1.3 datasets.

With the rapid development of 3D movie and light-field displays, there is a growing demand for stereo videos. However, generating high-quality stereo videos from planar videos remains a challenging task. Traditional depth-image-based rendering techniques struggle to effectively handle the problem of occlusion exposure, which occurs when the occluded contents become visible in other views. Recently, the single-view multiplane images (MPI) representation has shown promising performance for planar video stereoscopy. However, the MPI still lacks real details that are occluded in the current frame, resulting in blurry artifacts in occlusion exposure regions. In fact, planar videos can leverage complementary information from adjacent frames to predict a more complete scene representation for the current frame. Therefore, this paper extends the MPI from still frames to the temporal domain, introducing the temporal MPI (TMPI). By extracting complementary information from adjacent frames based on optical flow guidance, obscured regions in the current frame can be effectively repaired. Additionally, a new module called masked optical flow warping (MOFW) is introduced to improve the propagation of pixels along optical flow trajectories. Experimental results demonstrate that the proposed method can generate high-quality stereoscopic or light-field videos from a single view and reproduce better occluded details than other state-of-the-art (SOTA) methods. https://github.com/Dio3ding/TMPI

This work presents a novel method called dense projection for unsupervised anomaly detection (DPAD). The main idea is maximizing the local density of (normal) training data and then determining whether a test data is anomalous or not by evaluating its density. Specifically, DPAD uses a deep neural network to learn locally dense representations of normal data. Since density estimation is computationally expensive, we minimize the local distances of the representations in an iteratively reweighting manner, where the weights are updated adaptively and the parameters are regularized to avoid model collapse (all representations collapse to a single point). Compared with many state-of-the-art methods of anomaly detection, our DPAD does not rely on any assumption about the distribution or spatial structure of the normal data and representations. Moreover, we provide theoretical guarantees for the effectiveness of DPAD. The experiments show that our method DPAD is effective not only in traditional one-class classification problems but also in scenarios with complex normal data composed of multiple classes.

State-of-the-art video-text retrieval (VTR) methods typically involve fully fine-tuning a pre-trained model (e.g. CLIP) on specific datasets. However, this can result in significant storage costs in practical applications as a separate model per task must be stored. To address this issue, we present our pioneering work that enables parameter-efficient VTR using a pre-trained model, with only a small number of tunable parameters during training. Towards this goal, we propose a new method dubbed Multimodal Video Adapter (MV-Adapter) for efficiently transferring the knowledge in the pre-trained CLIP from image-text to video-text. Specifically, MV-Adapter utilizes bottleneck structures in both video and text branches, along with two novel components. The first is a Temporal Adaptation Module that is incorporated in the video branch to introduce global and local temporal contexts. We also train weights calibrations to adjust to dynamic variations across frames. The second is Cross Modality Tying that generates weights for video/text branches through sharing cross modality factors, for better aligning between modalities. Thanks to above innovations, MV-Adapter can achieve comparable or better performance than standard full fine-tuning with negligible parameters overhead. Notably, MV-Adapter consistently outperforms various competing methods in V2T/T2V tasks with large margins on five widely used VTR benchmarks (MSR-VTT, MSVD, LSMDC, DiDemo, and ActivityNet). Codes will be released.

Audio-driven talking-head synthesis has become a significant focus in the field of virtual human applications. However, existing methodologies face challenges in effectively synchronizing audio and video, especially in maintaining emotional consistency. Additionally, there is a notable inefficiency in leveraging emotional prompts to guide expression generation. To address these limitations, this paper introduces an Emotion Synchronized audio-driven Talking-head synthesis (EST) approach. The EST approach aims to enhance the emotion-agnostic talking-head models by enabling emotion control, and it incorporates a diffusion module to learn diverse latent rep-resentations. Furthermore, EST utilizes null-text embedding to align the latent code with emotional prompts. Additionally, a novel Sync Attention Block (SAB) is developed to broaden the spatial perceptual field, thus preventing the loss of critical information. Extensive experiments demonstrate the effectiveness of the EST method, showcasing state-of-the-art performance across widely-adopted datasets. Moreover, the EST approach exhibits exceptional generalization capabilities, even in scenarios where emotional training videos are unavailable.

Text-guided image generative diffusion models achieve fast development on the generation and editing of high-quality images. To extend such success to video editing, some efforts combining image generation with video editing have been made, which however only achieve inferior performance. We attribute it to two challenges: 1) different from the static image generation, it is tricky for dynamic video information to ensure the temporal fidelity of motion consistency across different frames; 2) the randomness of the frame generation process makes it hard to continuously retain the similar spatial fidelity for the original detailed features. In this paper, we propose a new high-fidelity diffusion model-based zero-shot text-guided video editing network, called HiFiVEditor, which aims to conduct effective video editing with high fidelity of the original video's detailed and dynamic information. Specifically, we propose a Spatial-Temporal Fidelity Block (STFB) that enables the model to restore the spatial features by enlarging the spatial perceptual field to avoid loss of important information, and capture more dynamic information between different frames by using all frames for preserving temporal consistency to achieve better temporal fidelity. In addition, we introduce Null-Text Embedding to create a soft text embedding to optimize the noise learning process, so that the latent noise can be aligned with the prompt. Furthermore, to tune the video style and render it more realistic, we employ a Prior-Guided Perceptual Loss to constrain the prediction results to avoid deviating from the original video style. Extensive experiments demonstrate the superior video editing capability compared to existing works.

Performing an object detection task after the restoration of a hazy image, or rather detecting with the network backbone directly, will result in the inclusion of information mixed with dehazing, which tends to interfere with detection performance. To address these issues, we propose a novel framework for detecting objects in dehazed images via Dual Cross-Stage Partial Learning (DCSP). Specifically, we introduce a Cross-Stage Partial (CSP) module for extracting clean feature information after dehazing. Secondly, to enhance data integrity, we employ a skip-input strategy to supplement information related to object detection features that may be lost during the dehazing task, while avoiding the gradient vanishing problem. In addition, CSP is also introduced to facilitate comprehensive learning of multiple feature representations. Finally, to avoid the inclusion of irrelevant dehazing information in detection, we apply a Ground-Truth Flow at detection network (at dark3), for fine feature information calibration. Additionally, we created a synthetic fog dataset to expand the training data for DCSP. Experimental results on both synthetic and real-world datasets demonstrate the effectiveness and accuracy of the proposed method. The code is available at https://github.com/zhaojinbiao/DCSP.

Neural networks have shown promising performance in collaborative filtering and matrix completion but the theoretical analysis is limited and there is still room for improvement in terms of the accuracy of recovering missing values. This paper presents a neuron-enhanced autoencoder matrix completion (AEMC-NE) method and applies it to collaborative filtering. Our AEMC-NE adds an element-wise autoencoder to each output of the main autoencoder to enhance the reconstruction capability. Thus it can adaptively learn an activation function for the output layer to approximate possibly complicated response functions in real data. We provide theoretical analysis for AEMC-NE as well as AEMC to investigate the generalization ability of autoencoder and deep learning in matrix completion, considering both missing completely at random and missing not at random. We show that the element-wise neural network has the potential to reduce the generalization error bound, the data sparsity can be useful, and the prediction performance is closely related to the difference between the numbers of variables and samples. The numerical results on synthetic data and benchmark datasets demonstrated the effectiveness of AEMC-NE in comparison to many baselines.

With the rapid development of high-bit-depth display devices, bit-depth expansion (BDE) algorithms that extend low-bit-depth images to high-bit-depth images have received increasing attention. Due to the sensitivity of bit-depth distortions to tiny numerical changes in the least significant bits, the nuanced degradation differences in the training process may lead to varying degradation data distributions, causing the trained models to overfit specific types of degradations. This paper focuses on the problem of blind video BDE, proposing a degradation prediction and embedding framework, and designing a video BDE network based on a recurrent structure and dual-frame alignment fusion. Experimental results demonstrate that the proposed model can outperform some state-of-the-art (SOTA) models in terms of banding artifact removal and color correction, avoiding overfitting to specific degradations and obtaining better generalization ability across multiple datasets. https://github.com/duanpanjun/BVBDE

Automatic generation of painting images is an interesting and difficult task, especially for regional traditional paintings with unique cultural styles while lacking large-scale training sets. In this paper, a hierarchical painting generation method is proposed, which can disentangle the generation of content and style. By mimicking the human painting process, the proposed method introduces multiple content blocks first and gradually generates image contents. In each block, a spatial self-modulation module is proposed to inject local details while preserving the global layout. After the preliminary generation of contents, a series of style blocks are presented to gradually adjust the artistic style. In the style block, an edge-oriented style-modulation module is proposed, which focuses on the lines and edges. In addition, edge adversarial training is used to further improve the quality of generated lines. To train and evaluate the proposed method, we construct datasets for five types of Chinese folk paintings. Experimental results demonstrate that the proposed method can generate high-quality and diverse painting images. More importantly, it can disentangle content and style sufficiently, so that the generation of specific contents or styles can be controlled freely. The datasets and source codes is available at https://github.com/Ritsu-mio/HPGN.

Underwater images often suffer from severe color distortion due to the challenging imaging environment. Underwater image enhancement (UIE) techniques have been developed to recover clear images, laying the foundation for various underwater research. However, existing UIE methods tend to produce fixed results without considering individual preferences for different color tones. And there is no dataset with ground truth (GT) in different tones. Therefore, we came up with the possibility of using the currently popular multimodal methods to control the color tone of enhanced images. This article proposes a method for generating underwater enhanced images with cold, warm, and normal tones using multimodal information supervision (MM-UIE). First, we leverage the relationship between text prompts and images to supervise the generation of cold or warm images. In addition, we introduce a 6-D color operator, which not only enhances the tone control of underwater images but also serves as a bridge between different tone images. Finally, we also found that multimodal supervision methods can not only control the color tone of underwater images but also improve the quality of underwater image generation. Experimental results demonstrate the superior performance of our method compared to state-of-the-art (SOTA) techniques. Our codes will be publicly available athttps://github.com/perseveranceLX/MM-UIE.

Stereo image dehazing aims to restore haze-free images by leveraging the complementary information contained in binocular images. Current methods primarily focus on designing image-level modules and pipelines to utilize complementary information between the left and right-view images. However, these image-level cross-view interactions overlook regional differences in haze concentration and stereo image disparity maps. Consequently, we propose a Progressive Stereo Image Dehazing Network via Cross-view Region Interaction, termed PSIDNet, which fully considers the internal characteristics and external manifestation of haze and disparity, and explicitly addresses the stereo image dehazing task by a regional-aware interactive mechanism. Specifically, we divide hazy images into regions and independently interact with left and right-view information at region levels, meaning weights are not shared across regional patches. This approach allows us to treat different regions with different priorities, i.e., concentrate on regional patches with heavier haze concentration and larger disparities, hence enabling more accurate restoration of hazy images. Furthermore, we introduce an effective cross-view region interactive block that extracts information based on the channel dimension of dual views and later adopts matrix multiplication to generate mutual attention maps based on the fused features. Extensive experiments on synthetic and real-scenario datasets demonstrate the efficacy of our method, compared to other related monocular and stereo image dehazing and restoration methods. Our code will be released publicly at https://github.com/Alvin2112/PSIDNet.

Today,fashion design can be readily performed by most people due to the rapid development of design tools. However, not everyone possesses the professional skills to produce an aesthetically pleasing design. In order to assist an inexperienced user during the design process, this research explored a new fashion-related disentanglement task, with the goal of creating novel fashion items with controllable attributes. The key idea is to develop a unified framework, called CTS-GAN, by disentangling the colors, textures, and shapes of fashion items simultaneously using a generative adversarial network (GAN). Specifically, we first introduced a fashion attribute encoder to decompose input fashion items into three latent spaces, i.e., color, texture, and shape. A fashion item pattern-making module (FIPM)-based generator was then proposed to control the corresponding parameters of color and texture in FIPMs independently and combine them with the shape features in order to accomplish the final generation of new fashion items. Furthermore, three independent pathways were introduced to extract the representations of color, texture, and shape in fashion items to optimize our CTS-GAN in an unsupervised manner. Extensive experimental results demonstrate the effectiveness of our CTS-GAN and suggest that it can generate diverse, novel fashion images by taking full advantage of the controllability of the colors, textures, and shapes of different fashion items.

In this article, we propose a new linear regression (LR)-based multiclass classification method, called discriminative regression with adaptive graph diffusion (DRAGD). Different from existing graph embedding-based LR methods, DRAGD introduces a new graph learning and embedding term, which explores the high-order structure information between four tuples, rather than conventional sample pairs to learn an intrinsic graph. Moreover, DRAGD provides a new way to simultaneously capture the local geometric structure and representation structure of data in one term. To enhance the discriminability of the transformation matrix, a retargeted learning approach is introduced. As a result of combining the above-mentioned techniques, DRAGD can flexibly explore more unsupervised information underlying the data and the label information to obtain the most discriminative transformation matrix for multiclass classification tasks. Experimental results on six well-known real-world databases and a synthetic database demonstrate that DRAGD is superior to the state-of-the-art LR methods.

The field of fashion compatibility learning has attracted great attention from both the academic and industrial communities in recent years. Many studies have been carried out for fashion compatibility prediction, collocated outfit recommendation, artificial intelligence (AI)-enabled compatible fashion design, and related topics. In particular, AI-enabled compatible fashion design can be used to synthesize compatible fashion items or outfits to improve the design experience for designers or the efficacy of recommendations for customers. However, previous generative models for collocated fashion synthesis have generally focused on the image-to-image translation between fashion items of upper and lower clothing. In this article, we propose a novel outfit generation framework, i.e., OutfitGAN, with the aim of synthesizing a set of complementary items to compose an entire outfit, given one extant fashion item and reference masks of target synthesized items. OutfitGAN includes a semantic alignment module (SAM), which is responsible for characterizing the mapping correspondence between the existing fashion items and the synthesized ones, to improve the quality of the synthesized images, and a collocation classification module (CCM), which is used to improve the compatibility of a synthesized outfit. To evaluate the performance of our proposed models, we built a large-scale dataset consisting of 20 000 fashion outfits. Extensive experimental results on this dataset show that our OutfitGAN can synthesize photo-realistic outfits and outperform the state-of-the-art methods in terms of similarity, authenticity, and compatibility measurements.

Synthesizing realistic images of fashion items which are compatible with given clothing images, as well as conditioning on multiple modalities, brings novel and exciting applications together with enormous economic potential. In this work, we propose a multi-modal collocation framework based on generative adversarial network (GAN) for synthesizing compatible clothing images. Given an input clothing item that consists of an image and a text description, our model works on synthesizing a clothing image which is compatible with the input clothing, as well as being guided by a given text description from the target domain. Specifically, a generator aims to synthesize realistic and collocated clothing images relying on image- and text-based latent representations learned from the source domain. An auxiliary text representation from the target domain is added for supervising the generation results. In addition, a multi-discriminator framework is carried out to determine compatibility between the generated clothing images and the input clothing images, as well as visual-semantic matching between the generated clothing images and the targeted textual information. Extensive quantitative and qualitative results demonstrate that our model substantially outperforms state-of-the-art methods in terms of authenticity, diversity, and visual-semantic similarity between image and text.

Deep learning based virtual try-on system has achieved some encouraging progress recently, but there still remain several big challenges that need to be solved, such as trying on arbitrary clothes of all types, trying on the clothes from one category to another and generating image-realistic results with few artifacts. To handle this issue, we in this article first collect a new dataset with all types of clothes, i.e., tops, bottoms, and whole clothes, each one has multiple categories with rich information of clothing characteristics such as patterns, logos, and other details. Based on this dataset, we then propose the Arbitrary Virtual Try-On Network (AVTON) that is utilized for all-type clothes, which can synthesize realistic try-on images by preserving and trading off characteristics of the target clothes and the reference person. Our approach includes three modules: (1) Limbs Prediction Module, which is utilized for predicting the human body parts by preserving the characteristics of the reference person. This is especially good for handling cross-category try-on task (e.g., long sleeves ↔ short sleeves or long pants ↔ skirts), where the exposed arms or legs with the skin colors and details can be reasonably predicted; (2) Improved Geometric Matching Module, which is designed to warp clothes according to the geometry of the target person. We improve the TPS based warping method with a compactly supported radial function (Wendland’s Ψ-function); (3) Trade-Off Fusion Module, which is to tradeoff the characteristics of the warped clothes and the reference person. This module is to make the generated try-on images look more natural and realistic based on a fine-tune symmetry of the network structure. Extensive simulations are conducted and our approach can achieve better performance compared with the state-of-the-art virtual try-on methods.

Landmark recognition stands as a prominent classification challenge within the domain of vision and perception, involving the identification and localization of landmarks in images. However, existing landmark recognition methods often fall short in delivering satisfactory performance. A critical issue lies in the lack of simultaneous exploration of local, regional, and global modeling. Intuitively, the local details, regional features, and global structure within an image, especially of a building, significantly contribute to landmark recognition. To address this issue, we propose a novel approach, termed Hierarchical Context Modeling Network (HCMNet), for landmark recognition. In our methodology, we introduce a Hierarchical Context Modeling Block (HCMB) designed with a triplet-branch structure to model image context in a hierarchical manner. These branches consist of the local branch, which leverages a convolutional layer to capture local details, the regional branch focused on extracting regional features through large-kernel asymmetrical convolutional layers, and the global branch aims at encompassing the broader context by incorporating channel attention. To enhance the modeling capacity, we assemble multiple HCMBs into the HCMNet, creating a hierarchical structure. Comprehensive experiments validate the efficacy of the proposed HCMNet for landmark recognition, showcasing its superior performance compared to current methods. The proposed hierarchical context modeling approach proves instrumental in capturing the intricate features essential for accurate landmark recognition.

Video super-resolution (VSR) on mobile devices aims to restore high-resolution frames from their low-resolution counterparts, satisfying the requirements of performance, FLOPs and latency. On one hand, partial feature processing, as a classic and acknowledged strategy, is developed in current studies to reach an appropriate trade-off between FLOPs and accuracy. However, the splitting of partial feature processing strategy are usually performed in a blind manner, thereby reducing the computational efficiency and performance gains. On the other hand, current methods for mobile platforms primarily treat VSR as an extension of single-image super-resolution to reduce model calculation and inference latency. However, lacking inter-frame information interaction in current methods results in a suboptimal latency and accuracy trade-off. To this end, we propose a novel architecture, termed Feature Aggregating Network with Inter-frame Interaction (FANI), a lightweight yet considering frame-wise correlation VSR network, which could achieve real-time inference while maintaining superior performance. Our FANI accepts adjacent multi-frame low-resolution images as input and generally consists of several fully-connection-embedded modules, i.e., Multi-stage Partial Feature Distillation (MPFD) for capturing multi-level feature representations. Moreover, considering the importance of inter-frame alignment, we further employ a tiny Attention-based Frame Alignment (AFA) module to promote inter-frame information flow and aggregation efficiently. Extensive experiments on the well-known dataset and real-world mobile device demonstrate the superiority of our proposed FANI, which means that our FANI could be well adapted to mobile devices and produce visually pleasing results.

Recent advances in single image super-resolution (SISR) have achieved remarkable performance through deep learning. However, the high computational cost hinders the deployment of SISR models on edge devices. Instead of proposing new SISR models, a new trend is emerging to improve network efficiency by reducing parameters, FLOPs, and inference time through slight modifications to the original models. However, recent methods usually focus on reducing only one of three metrics, i.e., FLOPs, parameters and inference time, which inevitably increases the other two metrics. In this paper, we propose a novel Adaptive Student Inference Network (ASIN) on popular SISR models, which aims at reducing FLOPs and inference time while maintaining the number of parameters and restoring clearer high-resolution images. Specifically, our ASIN divides a SISR model into three components (head, body and tail) and adopts various strategies for each part. For head and tail parts, to ensure the restored images contain more detailed information, a novel auxiliary Enhanced Teacher Network (ETNet) is designed, which is trained with the ground-truth images to obtain more prior knowledge to guide student network to extract more accurate textures using a new knowledge distillation method. For the body part, owing to the varying difficulties of the reconstructions in different regions, we propose an Adaptive Depth Predicted Module (ADPM) to dynamically shorten average depth of network to reduce the computational cost of overall network. Extensive experiments on two datasets demonstrate the effectiveness and state-of-the-art performance of our ASIN compared to its counterparts.

The human bodies are hierarchical structure and human pose estimation is highly dependent on the correlation between the keypoints of human body. Most existing CNN-based and Transformer-based methods perform well in the visual representation but lack the ability to learn the correlation between keypoints explicitly. To address this problem, we propose a human pose estimation method based on a hierarchical graph neural network with dynamic keypoint weight. The hierarchical graph neural network consists of three parts: the inter-layer process, hierarchical process, and iterative inference. The inter-layer process and hierarchical process learn the correlation between similar keypoints and further keypoints, respectively. And iterative inference is used to refine and aggregate more useful information between keypoints. Furthermore, the dynamic keypoint weight network gives each keypoint a different degree of importance. Extensive experiments demonstrate that our HGNPose-M and HGNPose-L achieve 76.5 AP $(\color{Red}{\uparrow 2.1})$ and 77.4 AP $(\color{Red}{\uparrow 1.6})$ on the COCO dataset respectively.

Low-Light Stereo Image Enhancement (LLSIE) tackles the challenge of improving the illumination and restoring the details in stereo images. However, existing deep learning-based LLSIE methods trained on high-resolution low-light images often exhibit sub-optimal performance when interacting with information from the left and right views. We find that this is because of: (1) the high computational cost arising from quadratic complexity, which hinders the enhancement model's ability to process high-resolution images; and (2) the limitations of conventional fusion strategies in previous work, which inadequately capture cross-view cues, resulting in weak feature representation and compromised detail recovery. To address these limitations, we propose a novel Dynamic Grouped Interaction Network (DGI-Net) to enhance illumination and recover more details while reducing the computational cost. Specifically, DGI-Net employs the U-Net structure, which effectively mitigates noise during the low-light enhancement. Furthermore, we design a Grouped Stereo Interaction Module (GSIM) with a grouping strategy to efficiently discover cross-view cues while minimizing computations. To dynamically fuse stereo information and fully exploit cross-view correlations, we also introduce a Dynamic Embedding Module (DEM) to establish dynamic connections between inter-view cues and intra-view features, which performs dynamic weight processing on cross-view cues to eliminate noise during fusion. For intra-view processing, we present a Diversity Enhanced Block (DEB) to extract multi-scale features, thereby improving diversity and feature representation. This multi-scale feature extraction also addresses low image contrast in dark lighting conditions. Experimental results demonstrate that DGI-Net outperforms current state-of-the-art methods in low-light stereo image enhancement.

With the popular of stereo video and free-viewpoint video, binocular and multi-view video enhancement has attracted increasing attention. Current binocular video enhancement methods mainly focus on stereo super-resolution. In this paper, we tend to discuss a new binocular video resolution and frame-rate enhancement scenario to fully utilize the cross-view complementary information. Specifically, one view is captured with high resolution (HR) and low frame-rate (LFR), while the other viewpoint records low resolution (LR) and high frame-rate (HFR) video. Then, a binocular video joint enhancement network, which adopts dual-branch structure with cross-view guidance, is proposed to jointly reconstruct HR and HFR stereo videos. The proposed framework can reduce the capture, storage, compression, and transmission cost of normal HR and HFR stereo videos. Compared with single-view super-resolution and video frame interpolation techniques, the proposed method can recover more realistic HR details and intermediate motion by using cross-view reference. Experimental results on stereo video datasets demonstrate the effectiveness of the proposed joint resolution and frame-rate enhancement framework.

Multi-stage architectures have exhibited efficacy in image dehazing, which usually decomposes a challenging task into multiple more tractable sub-tasks and progressively estimates latent hazy-free images. Despite the remarkable progress, existing methods still suffer from the following shortcomings: (1) limited exploration of frequency domain information; (2) insufficient information interaction; (3) severe feature redundancy. To remedy these issues, we propose a novel Mutual Information-driven Triple interaction Network (MITNet) based on spatial-frequency dual domain information and two-stage architecture. To be specific, the first stage, named amplitude-guided haze removal, aims to recover the amplitude spectrum of the hazy images for haze removal. And the second stage, named phase-guided structure refined, devotes to learning the transformation and refinement of the phase spectrum. To facilitate the information exchange between two stages, an Adaptive Triple Interaction Module (ATIM) is developed to simultaneously aggregate cross-domain, cross-scale, and cross-stage features, where the fused features are further used to generate content-adaptive dynamic filters so that applying them to enhance global context representation. In addition, we impose the mutual information minimization constraint on paired scale encoder and decoder features from both stages. Such an operation can effectively reduce information redundancy and enhance cross-stage feature complementarity. Extensive experiments on multiple public datasets exhibit that our MITNet performs superior performance with lower model complexity. The code and models are available at https://github.com/it-hao/MITNet.

Traditional fashion design typically requires the expertise of designers, which limits the involvement of ordinary users during the design process. While it would be desirable for users to participate in the preliminary design phase, their lack of basic design knowledge may render them too inexperienced to produce satisfactory designs. To improve the design efficiency for common users, we present a novel interactive fashion design framework based on generative adversarial network (GAN). This framework can assist users in designing fashion items by drawing only rough scribbles and providing simple fashion styles. Specifically, we propose a new cross-domain feature fusion encoder network that maps design image features from different domains into a series of style vectors which are then fed into a generator. We demonstrate that the learned style vectors can decouple the representations of cross-domain design elements and control the design results through scribbles and style images. Furthermore, we propose a method for rewriting our model with scribbles and style images, to allow designers to train our model more easily. To examine the effectiveness of our proposed model, we constructed a large-scale dataset containing 90,000 pairs of fashion item images. Experimental results show that our proposed method outperforms state-of-the-art methods and can effectively control cross-domain image features, suggesting the potential of our model for providing users with an intelligence-driven interactive design tool.

Transformer-based approaches to image captioning have shown great success by utilizing long-term dependency for visual embedding. However, their coarse long-term dependency, using the multi-head self-attention mechanism to capture the contextual interactions between the visual tokens on the time step and (or) embedded dimension, fail to distinguish fine-grained features of local partition. In this case, some similar features are captured, which leads to feature redundancy that decreases the performance. To respond to this issue, this paper proposes a novel image captioner embedding visual cross-partition dependency, dubbed CropCap. Specifically, the visual sequence generated from the Swin Transformer-based pre-embedding network is fed into the proposed cross-partition dependency module to refinedly model the interaction between partial representations on both the time step and embedded dimension. Furthermore, we formulaically reason the proposed cross-partition dependency, and theoretically prove its correctness. Extensive comparisons on the benchmark MS-COCO dataset demonstrated the effectiveness addressing the information redundancy issue, and verified the superior performance of our method.

The process of fashion design involves creative expression through various methods, including sketch drawing, brush painting, and choices of textures and colors, all of which are employed to characterize the originality and uniqueness of the designed fashion items. Despite recent advances in intelligence-driven fashion design, the complexity of the diverse elements of a fashion item, such as its texture, color and shape, which are associated with the semantic information conveyed, continues to present challenges in terms of generating high-quality fashion images as well as achieving a controllable editing process. To address this issue, we propose a unified framework, FashionDiff, that leverages the diverse elements in fashion items to generate new items. Initially, we collected a large number of fashion images with multiple categories and created pairwise data in terms of sketch and additional data, such as brush areas, textures, or colors. To eliminate semantic discrepancies between these pairwise datasets, we introduce a feature modulation fusion (FMFusion) process, which enables interactive communication among different images, allowing them to be fused into latent spaces characterized by different resolutions. In order to produce high-quality editable fashion images, we develop a generator based on a state-of-the-art diffusion model called FD-ControlNet, which integrates latent spaces into different layers of the generator to generate ready-to-wear fashion items. Qualitative and quantitative experimental results demonstrate the effectiveness of our proposed method, and suggest that our model can offer flexible control over the generated images in terms of sketches, brush areas, textures, and colors.

Texture constitutes the color and fabric of fashion items. Its choice in fashion items can directly express the personality and emotional state of a wearer. Despite the rapid development of intelligence-driven fashion design, it remains challenging to achieve independent control over texture without affecting other attributes, due to the highly intertwined nature of texture space. To accomplish fine-grained texture disentanglement, we propose InspirNET, an unsupervised disentangled generative adversarial framework, that manipulates textures in a fine-grained latent space so as to produce new textures effectively, aiming to broaden the range of fashion options available to common users with distinct textures as well as boosting designers' potential for fashion innovation and inspiration. Specifically, we first introduce an auto-fashion attribute encoder to map the input fashion item into texture and structure spaces. To achieve unsupervised fine-grained texture disentanglement, our model proposes a K-textures disentanglement module that decomposes the texture space into several orthogonal vectors, each of which is empowered to control an independent texture element. In particular, by employing an orthogonal eigenvector to interpolate with another, a multitude of new textures can be generated easily. Qualitative and quantitative experiments demonstrate that our InspirNET can effectively utilize decomposed orthogonal vectors to generate a wide range of fashion items with diverse textures. Our model exhibits superior performance over state-of-the-art methods in terms of maintaining the authenticity of texture transfer.

Low-light stereo image enhancement (LLSIE) aims at improving the visual quality of stereo images captured in dark conditions. However, existing methods have shown limited success in detail recovery and illumination adjustment. This can be attributed to two main factors: 1) insufficient single-scale inter-view interaction hinders the exploitation of valuable cross-view cues; 2) lacking long-range dependency leads to the inability to deal with the spatial long-range effects caused by illumination degradation. To address these limitations, we propose a novel LLSIE model named Decoupled Cross-Scale Cross-View Interaction Network (DCI-Net). Our model introduces a key component called the Decoupled Interaction Module (DIM) designed to promote sufficient dual-view information exchange. DIM decouples the dual-view information interaction by discovering multi-scale cross-view correlations and further exploring cross-scale information flow. Furthermore, we present Spatial-channel Information Mining Block (SIMB) for intra-view feature extraction, and the benefits are twofold. One is long-range dependency capture to build spatial long-range relationship, and the other is expanded channel information refinement that enhances information flow in the channel dimension. Extensive experiments conducted on Flickr1024, KITTI 2012, KITTI 2015, and Middlebury datasets show that our method obtains better illumination adjustment and detail recovery, and achieves SOTA performance compared to other related methods.

Outfit generation is a challenging task in the field of fashion technology, in which the aim is to create a collocated set of fashion items that complement a given set of items. Previous studies in this area have been limited to generating a unique set of fashion items based on a given set of items, without providing additional options to users. This lack of a diverse range of choices necessitates the development of a more versatile framework. However, when the task of generating collocated and diversified outfits is approached with multimodal image-to-image translation methods, it poses a challenging problem in terms of non-aligned image translation, which is hard to address with existing methods. In this research, we present FCBoost-Net, a new framework for outfit generation that leverages the power of pre-trained generative models to produce multiple collocated and diversified outfits. Initially, FCBoost-Net randomly synthesizes multiple sets of fashion items, and the compatibility of the synthesized sets is then improved in several rounds using a novel fashion compatibility booster. This approach was inspired by boosting algorithms and allows the performance to be gradually improved in multiple steps. Empirical evidence indicates that the proposed strategy can improve the fashion compatibility of randomly synthesized fashion items as well as maintain their diversity. Extensive experiments confirm the effectiveness of our proposed framework with respect to visual authenticity, diversity, and fashion compatibility.

Virtual devices based on device emulation have been widely used in lab research of mobile app testing for their efficiency and low cost. However, it remains controversial to use virtual devices for app testing in industry, given the inherent difficulties of high-fidelity emulation across diverse mobile systems and devices. Hence, mobile app companies still rely on physical device farms or services like AWS Device Farm.

Graph convolutional networks (GCN) with one or two hidden layers have been widely used in handling graph data that are prevalent in various disciplines. Many studies showed that the gain of making GCNs deeper is tiny or even negative. This implies that the complexity of graph data is often limited and shallow models are often sufficient to extract expressive features for various tasks such as node classification. Therefore, in this work, we present a framework called graph convolutional kernel machine (GCKM) for graph-based machine learning. GCKMs are built upon kernel functions integrated with graph convolution. An example is the graph convolutional kernel support vector machine (GCKSVM) for node classification, for which we analyze the generalization error bound and discuss the impact of the graph structure. Compared to GCNs, GCKMs require much less effort in architecture design, hyperparameter tuning, and optimization. More importantly, GCKMs are guaranteed to obtain globally optimal solutions and have strong generalization ability and high interpretability. GCKMs are composable, can be extended to large-scale data, and are applicable to various tasks (e.g., node or graph classification, clustering, feature extraction, dimensionality reduction). The numerical results on benchmark datasets show that, besides the aforementioned advantages, GCKMs have at least competitive accuracy compared to GCNs.

Recognizing the unseen combinations of action and different objects, namely (zero-shot) compositional action recognition, is extremely challenging for conventional action recognition algorithms in real-world applications. Previous methods focus on enhancing the dynamic clues of objects that appear in the scene by building region features or tracklet embedding from ground-truths or detected bounding boxes. These methods rely heavily on manual annotation or the quality of detectors, which are inflexible for practical applications. In this work, we aim to mining the temporal clues from moving objects or hands without explicit supervision. Thus, we propose a novel Motion Stimulation (MS) block, which is specifically designed to mine dynamic clues of the local regions autonomously from adjacent frames. Furthermore, MS consists of the following three steps: motion feature extraction, motion feature recalibration, and action-centric excitation. The proposed MS block can be directly and conveniently integrated into existing video backbones to enhance the ability of compositional generalization for action recognition algorithms. Extensive experimental results on three action recognition datasets, the Something-Else, IKEA-Assembly and EPIC-KITCHENS datasets, indicate the effectiveness and interpretability of our MS block.

Zero-shot learning (ZSL) aims to recognize unknown categories that are unavailable during training. Recently, generative models have shown the potential to address this challenging problem by synthesizing unseen features conditioned on semantic embeddings such as attributes. However, unidirectional generative models cannot guarantee the effective coupling between visual and semantic spaces. To this end, we propose a visual-semantic aligned bidirectional network with cycle consistency to alleviate the gap between these two spaces, generating unseen features of high quality. More importantly, we incorporate two carefully designed strategies into our bidirectional framework to improve the overall ZSL performance. Specifically, we enhance the intra-domain class divergence in both visual and semantic spaces, and in the meantime, mitigate the inter-domain shift to preserve seen-unseen domain discrimination. Experimental results on four standard benchmarks show the superiority of our framework over existing state-of-the-art methods under both conventional and generalized ZSL settings.

The traditional fashion industry is heavily dependent on designers whose talent and vision have a significant impact on their innovative designs. Through taking advantage of recent advances in image-to-image translation by generative adversarial networks (GANs), marked improvement in designers’ efficiency is now possible. Considering both randomness and controllability in the design process, this article presents a novel artificial intelligence (AI)-based framework for fashion design. Under this framework, a sketch-generation module which is based on latent space is firstly introduced for designing various sketches. Secondly, a rendering-generation module is proposed to learn mapping between textures and sketches to complete the task of fashion design. In order to achieve effectiveness in synthesizing semantic-aware textures on sketches, a multi-conditional feature interaction module is developed in the rendering-generation model. Moreover, two different training schemes are introduced to optimize both the sketch-generation module and the rendering-generation module. In order to evaluate the performance of our proposed models, we built a large-scale dataset which consists of 115,584 pairs of fashion item images. Experimental results demonstrate the effectiveness of our proposed method, and indicate that our model can facilitate designers’ design process by taking full advantage of the controllability of different conditions (e.g., sketch and texture) and the randomness of latent space.

Conventional multiview clustering seeks to partition data into respective groups based on the assumption that all views are fully observed. However, in practical applications, such as disease diagnosis, multimedia analysis, and recommendation system, it is common to observe that not all views of samples are available in many cases, which leads to the failure of the conventional multiview clustering methods. Clustering on such incomplete multiview data is referred to as incomplete multiview clustering (IMC). In view of the promising application prospects, the research of IMC has noticeable advances in recent years. However, there is no survey to summarize the current progresses and point out the future research directions. To this end, we review the recent studies of IMC. Importantly, we provide some frameworks to unify the corresponding IMC methods and make an in-depth comparative analysis for some representative methods from theoretical and experimental perspectives. Finally, some open problems in the IMC field are offered for researchers. The related codes are released athttps://github.com/DarrenZZhang/Survey_IMC.

Clothes attribution prediction is the key technology for users to automatically describe clothing characteristics. Most current methods are first to detect the multiple clothes, and then crop out the clothes and feed to a certain network for clothes attribution prediction. But this two-stage approach is time- and resource- consuming; on the other hand, one-stage approach can provide an effective and efficient solution by integrating clothes detection and attribution prediction into an end-to-end framework. But the one-stage approach tends to explore anchor-based detectors causing high sensitivity to the hyperparameters and high computational complexity from dense anchors. In addition, it may also confront with optimization contradiction problem in the training procedure, as the clothes detection and attribution prediction branches demand diverse optimization. In this work, to handle the above problems, we aim to develop an end-to-end anchor-free framework by involving an additional branch for joint clothes detection and attribution prediction. To handle the optimization contradiction in two branches, we encode the backbone feature map as pixel-level dense queries and decode them via deformable transformer as the output features that are fed into detection and prediction branches, respectively. In this way, the features of detection and prediction branches can be decoupled and the optimization contradiction can be naturally solved. To further enhance the prediction accuracy, we in the prediction branch also develop a special attention strategy and loss function to adaptively integrate the peer attribution relationships into feature learning as well as to avoid mutual suppression for hierarchical attributions. Extensive simulation results verify the effectiveness of the proposed work.

Recently deep learning methods have shown significant progress in data clustering tasks. Deep clustering methods (including distance-based methods and subspace-based methods) integrate clustering and feature learning into a unified framework, where there is a mutual promotion between clustering and representation. However, deep subspace clustering methods are usually in the framework of self-expressive model and hence have quadratic time and space complexities, which prevents their applications in large-scale clustering and real-time clustering. In this paper, we propose a new mechanism for deep clustering. We aim to learn the subspace bases from deep representation in an iterative refining manner while the refined subspace bases help learning the representation of the deep neural networks in return. The proposed method is out of the self-expressive framework, scales to the sample size linearly, and is applicable to arbitrarily large datasets and online clustering scenarios. More importantly, the clustering accuracy of the proposed method is much higher than its competitors. Extensive comparison studies with state-of-the-art clustering approaches on benchmark datasets demonstrate the superiority of the proposed method.

Low-light image enhancement (LLIE) explores how to refine the illumination and obtain natural normal-light images. Current LLIE methods mainly focus on improving the illumination, but do not consider the color consistency by reasonably incorporating color information into the LLIE process. As a result, color difference usually exists between the enhanced image and ground-truth. To address this issue, we propose a new deep color consistent network termed DCC-Net to retain the color consistency for LLIE. A new “divide and conquer” collaborative strategy is presented, which can jointly preserve color information and enhance the illumination. Specifically, the decoupling strategy of our DCC-Net decouples each color image into two main components, i.e., gray image plus color histogram. Gray image is used to generate reasonable structures and textures, and the color histogram is beneficial for preserving the color consistency. That is, they both are utilized to complete the LLIE task collaboratively. To match the color and content features, and reduce the color consistency gap between enhanced image and ground-truth, we also design a new pyramid color embedding (PCE) module, which can better embed color information into the LLIE process. Extensive experiments on six real datasets show that the enhanced images of our DCC-Net are more natural and colorful, and perform favorably against the state-of-the-art methods.

To obtain lower inference latency and less memory footprint of deep neural networks, model quantization has been widely employed in deep model deployment, by converting the floating points to low-precision integers. However, previous methods (such as quantization aware training and post training quantization) require original data for the fine-tuning or calibration of quantized model, which makes them inapplicable to the cases that original data are not accessed due to privacy or security. This gives birth to the data-free quantization method with synthetic data generation. While current data-free quantization methods still suffer from severe performance degradation when quantizing a model into lower bit, caused by the low inter-class separability of semantic features. To this end, we propose a new and effective data-free quantization method termed ClusterQ, which utilizes the feature distribution alignment for synthetic data generation. To obtain high inter-class separability of semantic features, we cluster and align the feature distribution statistics to imitate the distribution of real data, so that the performance degradation is alleviated. Moreover, we incorporate the diversity enhancement to solve class-wise mode collapse. We also employ the exponential moving average to update the centroid of each cluster for further feature distribution improvement. Extensive experiments based on different deep models (e.g., ResNet-18 and MobileNet-V2) over the ImageNet dataset demonstrate that our proposed ClusterQ model obtains state-of-the-art performance.

Image captioning is still a challenging task aiming at describing the contents of image by words. Current image caption methods usually assume the object relation to be important if the semantic and spatial geometric relationships between objects are close and large, but the relations meeting this assumption are not necessarily important to describe the contents of image in a fine-grained way. That is, the importance of fine-grained object relations is not properly taken into account. Besides, current Transformer based image caption models also fail to consider the importance of fine-grained objects, since they generate all the words of a sentence at one time, which cannot Figure out which objects are more important and vice versa. In this paper, we propose a novel Fine-grained Adaptive Object Transformer (FineFormer) network, which can jointly discover the importance of fine-grained objects and object relations for image captioning. Specifically, a new concept of adaptive soft-foreground attention is proposed to highlight the fine-grained objects dominating the descriptive contents. To characterize and calculate the important relations between fine-grained objects, we also propose an adaptive object relation attention to refine the object relation from the generation process of relation. As such, FineFormer can describe the contents of image more accurately, by reducing the interference of unimportant objects in the background. Extensive experiments on the highly-competitive MS-COCO dataset demonstrated the superiority of our FineFormer.

Convolutional Neural Networks (CNNs) are powerful for image representation, but the convolution operation may be influenced and degraded by the included noise, and the deep features may not be fully learned. In this paper, we propose a new encoder-decoder based image restoration network, termed Robust Low-Rank Convolution Network with Feature Denoising (LRCnet). LRCnet presents a novel low-rank convolution (LR-Conv) for image representation, and a residual dense connection (RDC) for feature fusion between encoding and decoding. Different from directly splitting convolution into ordinary convolution and mirror convolution as existing work, LR-Conv deploys a feature denoising module after the ordinary convolution to remove noise for mirror convolution. A low-rank embedding process is then used to project the convolutional features into a robust low-rank subspace, which can retain the local geometry of input signal to some extent and separate the signal and noise by finding low-rank structure of features to reduce the impact of noise on convolution. Besides, most networks increase the depth of network simply to obtain deep information and lack of effective connections to fuse the multilevel features, which may not fully discover the deep features in various layers. Thus, we design a residual dense connection with a channel attention to connect multilevel feature effectively to obtain more useful information to enhance the data representation. Extensive experiments on several datasets verified the effectiveness of LRCnet for image denoising.

Synchronous Rain streaks and Raindrops Removal (SR^3) is a hard and challenging task, since rain streaks and raindrops are two wildly divergent real-world phenomena with different optical properties and mathematical distributions. As such, most existing data-driven deep Singe Image Deraining (SID) methods only focus on one of them. Although there are only a few existing SR^3 methods, they still suffer from blur textures and unknown noise in reality due to weak robustness and generalization ability. In this paper, we will propose a new and universal SID model with novel modules, termed Robust Attention Deraining Network (RadNet), with strong robustness and generalization ability that are reflected in two main aspects. (1) RadNet can restore different rain degenerations, including raindrops, rain streaks, or both; (2) RadNet can adapt to different data strategies, including single-type, superimposed-type, and blended-type. The generalization ability is also demonstrated by the performance of dealing with real rain images. Specifically, we first design a lightweight and robust attention module (RAM) with a universal attention mechanism for coarse rain removal, and then present a new deep refining module (DRM) with multi-scale blocks for precise rain removal. To solve the inconsistent labels of real scenario data, we also introduce a flow & warp module (FWM) into the network, which can greatly improve the performance of real scenario data via optical flow prediction and alignment. The whole process is unified in a network to ensure sufficient robustness and strong generalization ability. We evaluated the performance of our method under a variety of data strategies, and extensive experiments demonstrated that our RadNet could outperform other state-of-the-art SID methods.

Data-driven single image deraining (SID) models have achieved greater progress by simulations, but there is still a large gap between current deraining performance and practical high-level applications, since high-level semantic information is usually neglected in current studies. Although few studies jointly considered high-level tasks (e.g., segmentation) to enable the model to learn more high-level information, there are two obvious shortcomings. First, they require the segmentation labels for training, limiting their operations on other datasets without high-level labels. Second, high- and low-level information are not fully interacted, hence having limited improvement in both deraining and segmentation tasks. In this paper, we propose a Semantic Guided Interactive Network (SGINet), which considers the sufficient interaction between SID and semantic segmentation using a three-stage deraining manner, i.e., coarse deraining, semantic information extraction, and semantics guided deraining. Specifically, a Full Resolution Module (FRM) without down-/up-sampling is proposed to predict the coarse deraining images without context damage. Then, a Segmentation Extracting Module (SEM) is designed to extract accurate semantic information. We also develop a novel contrastive semantic discovery (CSD) loss, which can instruct the process of semantic segmentation without real semantic segmentation labels. Finally, a triple-direction U-net-based Semantic Interaction Module (SIM) takes advantage of the coarse deraining images and semantic information for fully interacting low-level with high-level tasks. Extensive simulations on the newly-constructed complex datasets Cityscapes_syn and Cityscapes_real demonstrated that our model could obtain more promising results. Overall, our SGINet achieved SOTA deraining and segmentation performance in both simulation and real-scenario data, compared with other representative SID methods.

Blind image deblurring (BID) remains a challenging and significant task. Benefiting from the strong fitting ability of deep learning, paired data-driven supervised BID methods have obtained great progress. However, paired data are usually synthesized by hand, and the realistic blurs are more complex than synthetic ones, which makes the supervised methods inept at modeling realistic blurs and hinders real-world applications. As such, unsupervised deep BID methods without paired data offer certain advantages, but current methods still suffer from some drawbacks, e.g., bulky model size, long inference time, and strict image resolution and domain requirements. In this paper, we propose a lightweight and real-time unsupervised BID baseline, termed Frequency-domain Contrastive Loss Constrained Lightweight CycleGAN (shortly, FCL-GAN), with attractive properties, i.e., no image domain limitation, no image resolution limitation, 25x lighter than SOTA, and 5x faster than SOTA. To guarantee the lightweight property and performance superiority, two new collaboration units called lightweight domain conversion unit (LDCU) and parameter-free frequency-domain contrastive unit (PFCU) are designed. LDCU mainly implements inter-domain conversion in lightweight manner. PFCU further explores the similarity measure, external difference and internal connection between the blurred domain and sharp domain images in frequency domain, without involving extra parameters. Extensive experiments on several image datasets demonstrate the effectiveness of our FCL-GAN in terms of performance, model size and inference time.

Blind image deblurring is still a challenging problem due to the inherent ill-posed properties. To improve the deblurring performance, many supervised methods have been proposed. However, obtaining labeled samples from a specific distribution (or a domain) is usually expensive, and the data-driven training-based model also cannot be generalized to the blurry images in all domains. These challenges have given birth to certain unsupervised deblurring methods. However, there is a great chromatic aberration between the latent and original images, directly degrading the performance. In this paper, we therefore propose a novel unsupervised color retention network termed CRNet to perform blind motion deblurring. In addition, new concepts of blur offset estimation and adaptive blur correction are proposed to retain the color information when deblurring. As a result, unlike the previous studies, CRNet does not learn a mapping directly from the blurry image to the restored latent image, but from the blurry image to a motion offset. An adaptive blur correction operation is then performed on the blurry image to restore the latent image, thereby retaining the color information of the original image to the greatest extent. To further effectively retain the color information and extract the blur information, we also propose a new module called pyramid global blur feature perception (PGBFP). To quantitatively prove the effectiveness of our network in color retention, we propose a novel chromatic aberration quantization metrics in line with the human perception. Extensive quantitative and visualization experiments show that CRNet can obtain the state-of-the-art performance in unsupervised deblurring tasks.

Deep image inpainting methods have improved the inpainting performance greatly due to the powerful representation ability of deep learning. However, current deep inpainting networks still tend to produce unreasonable structures and blurry textures due to the ill-posed properties of the task, i.e., image inpainting is still a challenging topic. In this paper, we therefore propose a novel deep multi-resolution mutual learning (DMRML) strategy, which can fully explore the information from various resolutions. Specifically, we design a new image inpainting network, termed multi-resolution mutual network (MRM-Net), which takes the damaged images of different resolutions as input, then excavates and exploits the correlation among different resolutions to guide the image inpainting process. Technically, we designs two new modules called multi-resolution information interaction (MRII) and adaptive content enhancement (ACE). MRII aims at discovering the correlation of multiple resolutions and exchanging information, and ACE focuses on enhancing the contents using the interacted features. Note that we also present an memory preservation mechanism (MPM) to prevent from the information loss with the increasing layers. Extensive experiments on Paris Street View, Places2 and CelebA-HQ datasets demonstrate that our proposed MRM-Net can effectively recover the textures and structures, and performs favorably against other state-of-the-art methods.

The performance of spectral clustering heavily relies on the quality of affinity matrix. A variety of affinity-matrix-construction (AMC) methods have been proposed but they have hyperparameters to determine beforehand, which requires strong experience and leads to difficulty in real applications, especially when the inter-cluster similarity is high and/or the dataset is large. In addition, we often need to choose different AMC methods for different datasets, which still depends on experience. To solve these two challenging problems, in this paper, we present a simple yet effective method for automated spectral clustering. First, we propose to find the most reliable affinity matrix via grid search or Bayesian optimization among a set of candidates given by different AMC methods with different hyperparameters, where the reliability is quantified by the \textit{relative-eigen-gap} of graph Laplacian introduced in this paper. Second, we propose a fast and accurate AMC method based on least squares representation and thresholding and prove its effectiveness theoretically. Finally, we provide a large-scale extension for the automated spectral clustering method, of which the time complexity is linear with the number of data points. Extensive experiments of natural image clustering show that our method is more versatile, accurate, and efficient than baseline methods.

Multiview learning (MVL), which enhances the learners' performance by coordinating complementarity and consistency among different views, has attracted much attention. The multiview generalized eigenvalue proximal support vector machine (MvGSVM) is a recently proposed effective binary classification method, which introduces the concept of MVL into the classical generalized eigenvalue proximal support vector machine (GEPSVM). However, this approach cannot guarantee good classification performance and robustness yet. In this article, we develop multiview robust double-sided twin SVM (MvRDTSVM) with SVM-type problems, which introduces a set of double-sided constraints into the proposed model to promote classification performance. To improve the robustness of MvRDTSVM against outliers, we take L1-norm as the distance metric. Also, a fast version of MvRDTSVM (called MvFRDTSVM) is further presented. The reformulated problems are complex, and solving them are very challenging. As one of the main contributions of this article, we design two effective iterative algorithms to optimize the proposed nonconvex problems and then conduct theoretical analysis on the algorithms. The experimental results verify the effectiveness of our proposed methods.

In this paper, we propose a novel method, referred to as incomplete multi-view tensor spectral clustering with missing-view inferring (IMVTSC-MVI) to address the challenging multi-view clustering problem with missing views. Different from the existing methods which commonly focus on exploring the certain information of the available views while ignoring both of the hidden information of the missing views and the intra-view information of data, IMVTSC-MVI seeks to recover the missing views and explore the full information of such recovered views and available views for data clustering. In particular, IMVTSC-MVI incorporates the feature space based missing-view inferring and manifold space based similarity graph learning into a unified framework. In such a way, IMVTSC-MVI allows these two learning tasks to facilitate each other and can well explore the hidden information of the missing views. Moreover, IMVTSC-MVI introduces the low-rank tensor constraint to capture the high-order correlations of multiple views. Experimental results on several datasets demonstrate the effectiveness of IMVTSC-MVI for incomplete multi-view clustering.

In this paper, we technically propose an enriched prior guided framework, called Dual-constrained Deep Semi-Supervised Coupled Factorization Network (DS2CF-Net), for discovering hierarchical coupled data representation. To extract hidden deep features, DS2CF-Net is formulated as a partial-label and geometrical structure-constrained framework. Specifically, DS2CF-Net designs a deep factorization architecture using multilayers of linear transformations, which can coupled update both the basis vectors and new representations in each layer. To enable learned deep representations and coefficients to be discriminative, we also consider enriching the supervised prior by joint deep coefficients-based label prediction and then incorporate the enriched prior information as additional label and structure constraints. The label constraint can enable the intra-class samples to have same coordinate in feature space, and the structure constraint forces the coefficients in each layer to be block-diagonal so that the enriched prior using the self-expressive label propagation are more accurate. Our network also integrates the adaptive dual-graph learning to retain the local structures of both data and feature manifolds in each layer. Extensive experiments on image datasets demonstrate the effectiveness of DS2CF-Net for representation learning and clustering.

Deep neural network (DNN) compression can reduce the memory footprint of deep networks effectively, so that the deep model can be deployed on the portable devices. However, most of the existing model compression methods cost lots of time, e.g., vector quantization or pruning, which makes them inept to the real-world applications that need fast online computation. In this paper, we therefore explore how to accelerate the model compression process by reducing the computation cost. Then, we propose a new deep model compression method, termed Dictionary Pair-based Data-Free Fast DNN Compression, which aims at reducing the memory consumption of DNNs without extra training and can greatly improve the compression efficiency. Specifically, our proposed method performs tensor decomposition on the DNN model with a fast dictionary pair learning-based reconstruction approach, which can be deployed on different layers (e.g., convolution and fully-connection layers). Given a pre-trained DNN model, we first divide the parameters (i.e., weights) of each layer into a series of partitions for dictionary pair-based fast reconstruction, which can potentially discover more fine-grained information and provide the possibility for parallel model compression. Then, dictionaries of less memory occupation are learned to reconstruct the weights. Extensive experiments on popular DNNs (i.e., VGG-16, ResNet-18 and ResNet-50) showed that our proposed weight compression method can significantly reduce the memory footprint and speed up the compression process, with less performance loss.

Convolutional Neural Networks (CNNs)-guided deep models have obtained impressive performance for image representation, however the representation ability may still be restricted and usually needs more epochs to make the model converge in training, due to the useful information loss during the convolution and pooling operations. We therefore propose a general feature recovery layer, termed Low-rank Deep Feature Recovery (LDFR), to enhance the representation ability of the convolutional features by seamlessly integrating low-rank recovery into CNNs, which can be easily extended to all existing CNNs-based models. To be specific, to recover the lost information during the convolution operation, LDFR aims at learning the low-rank projections to embed the feature maps onto a low-rank subspace based on some selected informative convolutional feature maps. Such low-rank recovery operation can ensure all convolutional feature maps to be reconstructed easily to recover the underlying subspace with more useful and detailed information discovered, e.g., the strokes of characters or the texture information of clothes can be enhanced after LDFR. In addition, to make the learnt low-rank subspaces more powerful for feature recovery, we design a fusion strategy to obtain a generalized subspace, which averages over all learnt sub-spaces in each LDFR layer, so that the convolutional feature maps in test phase can be recovered effectively via low-rank embedding. Extensive results on several image datasets show that existing CNNs-based models equipped with our LDFR layer can obtain better performance.

Real-world data in emerging applications may suffer from highly-skewed class imbalanced distribution, however how to deal with this kind of problem appropriately through deep learning needs further investigation. In this paper, we mainly propose a novel cross-entropy based loss function, referred to as Additive Scale Loss (ASL), for deep representation learning and imbalanced classification. To deal with the class imbalanced problem, ASL aims at increasing the loss in case of misclassification, which can avoid the superimposed loss values caused by the large amount of easily classified data in the unbalanced database to dominate the loss value of misclassified data. Moreover, in real-world applications, one data source may be used for multiple scenarios, such as classification and embedding learning, however training two separable models to handle these problems is costly, especially in deep learning area. To tackle this issue, we present and integrate a discriminative inter-class separation term into ASL, and propose a discriminative ASL (D-ASL), which can not only improve the classification performance, but also obtain discriminative representations simultaneously. The discriminative inter-class separation term is general, and can be easily integrated to other loss functions, such as CE and FL, as the byproducts. Finally, a new deep convolutional neural network equipped with D-ASL and a fully-connected (FC) layer is proposed, which can classify the imbalanced image data and obtain the discriminative representations at the same time. Extensive experimental results verified the superior performance of our method.

Deep subspace clustering (DSC) with the auto-encoder and self-expression layer is of great concern due to encouraging performance. However, existing methods usually adopt a “single-task” strategy based on a single dataset, without considering other related tasks or data. As such, they cannot discover other useful information to improve the clustering task. Besides, the local structure preservation of the latent codes in mapping is usually ignored. In this paper, we therefore present an effective “multi-task” strategy via the self-supervised data augmentation, and propose a new end-to-end trainable Triplet Deep Subspace Clustering Network (TDSC-net). Specifically, TDSC-net firstly generates triplet data (i.e., anchor, positive and negative data) from input data by a spectral clustering module and a self-supervised data augmentation module. This can enable it to inherit the merits of self-supervised learning and multitask learning implicitly. After that, TDSC-net builds a triplet deep autoencoder network with a self-expression layer, which takes the triplet data as input, where they share the common network layers (i.e., autoencoder and self-expression layers) over the triple tasks for complementary learning and mutual supervision. A triplet loss is also included to retain the local information of deep latent codes, which will also benefit the self-expression. Furthermore, TDSC-net separates the self-expression layer from decoding process to improve the efficiency of reconstruction. Extensive results on several public datasets demonstrate the effectiveness of our triplet-task DSC strategy.

Although supervised single image deraining (SID) have obtained impressive results, they still cannot obtain satisfactory results on real images for the weak generalization of rain removal capacity. In this paper, we mainly discuss the semi-supervised SID and propose a new GAN-based deraining network called Semi-DerainGAN, which can use both synthetic and real data in a uniform network based on two supervised and unsupervised processes. For this task, a semi-supervised rain streak learner termed SSRML sharing the same parameters of both processes is derived, which makes the real images contribute more rain streak information, so that the resulted model has a strong generalization power to the real SID task. We also contribute a new real-world rain image dataset called Real200 to alleviate the difference between both synthetic and real image domains. Extensive results on public datasets show that our model can obtain competitive results, especially on the real rain images.

Fisher-regularized support vector machine (Fish-erSVM) can approximatively fulfill the Fisher criterion and obtain good statistical separability, which is a combined method of the support vector machine and Fisher discriminant analysis. However, the hinge loss function is related to the shortest distance between two-class sets, and FisherSVM may be hence sensitive to noise. To remedy it, the pinball loss function, which is related to the quantile distance, is introduced into FisherSVM and then a Fisher-regularized support vector machine with pinball loss function (Pin-FisherSVM) is proposed, which combines the noise insensitivity of the pinball loss function with the statistical separability of FisherSVM well. Pin-FisherSVM can be cast as a quadratic programming, which results a globally optimal solution. Experimental results on artificial and real-world datasets demonstrate that our proposed method is insensitive to label noise or feature noise. Compared to FisherSVM, the proposed Pin-FisherSVM has the same computational complexity and exhibits superior noise insensitivity.

Deep learning based inpainting methods have obtained promising performance for image restoration, however current image inpainting methods still tend to produce unreasonable structures and blurry textures when processing the damaged images with heavy corruptions. In this paper, we propose a new image inpainting method termed Global Context Modeling Network (GCM-Net). By capturing the global contextual information, GCM-Net can potentially improve the performance of recovering the missing region in the damaged images with irregular masks. To be specific, we first use four convolution layers to extract the shadow features. Then, we design a progressive multi-scale fusion block termed PMSFB to extract and fuse the multi-scale features for obtaining local features. Besides, a dense context extraction (DCE) module is also designed to aggregate the local features extracted by PMSFBs. To improve the information flow, a channel attention guided residual learning module is deployed in both the DCE and PMSFB, which can reweight the learned residual features and refine the extracted information. To capture more global contextual information and enhance the representation ability, a coordinate context attention (CCA) based module is also presented. Finally, the extracted features with rich information are decoded as the image inpainting result. Extensive results on the Paris Street View, Places2 and CelebA-HQ datasets demonstrate that our method can better recover the structures and textures, and deliver significant improvements, compared with some related inpainting methods.

An important underlying assumption that guides the success of the existing multiview learning algorithms is the full observation of the multiview data. However, such rigorous precondition clearly violates the common-sense knowledge in practical applications, where in most cases, only incomplete fractions of the multiview data are given. The presence of the incomplete settings generally disables the conventional multiview clustering methods. In this article, we propose a simple but effective incomplete multiview clustering (IMC) framework, which simultaneously considers the local geometric information and the unbalanced discriminating powers of these incomplete multiview observations. Specifically, a novel graph-regularized matrix factorization model, on the one hand, is developed to preserve the local geometric similarities of the learned common representations from different views. On the other hand, the semantic consistency constraint is introduced to stimulate these view-specific representations toward a unified discriminative representation. Moreover, the importance of different views is adaptively determined to reduce the negative influence of the unbalanced incomplete views. Furthermore, an efficient learning algorithm is proposed to solve the resulting optimization problem. Extensive experimental results performed on several incomplete multiview datasets demonstrate that the proposed method can achieve superior clustering performance in comparison with some state-of-the-art multiview learning methods.

Single Image Deraining (SID) is a relatively new and still challenging topic in emerging vision applications, and most of the recently emerged deraining methods use the supervised manner depending on the ground-truth (i.e., using paired data). However, in practice it is rather common to encounter unpaired images in real deraining task. In such cases, how to remove the rain streaks in an unsupervised way will be a challenging task due to lack of constraints between images and hence suffering from low-quality restoration results. In this paper, we therefore explore the unsupervised SID issue using unpaired data, and propose a new unsupervised framework termed DerainCycleGAN for single image rain removal and generation, which can fully utilize the constrained transfer learning ability and circulatory structures of CycleGAN. In addition, we design an unsupervised rain attentive detector (UARD) for enhancing the rain information detection by paying attention to both rainy and rain-free images. Besides, we also contribute a new synthetic way of generating the rain streak information, which is different from the previous ones. Specifically, since the generated rain streaks have diverse shapes and directions, existing derianing methods trained on the generated rainy image by this way can perform much better for processing real rainy images. Extensive experimental results on synthetic and real datasets show that our DerainCycleGAN is superior to current unsupervised and semi-supervised methods, and is also highly competitive to the fully-supervised ones.

Concept Factorization (CF) and its variants may produce inaccurate representation and clustering results due to the sensitivity to noise, hard constraint on the reconstruction error, and pre-obtained approximate similarities. To improve the representation ability, a novel unsupervised Robust Flexible Auto-weighted Local-coordinate Concept Factorization (RFA-LCF) framework is proposed for clustering high-dimensional data. Specifically, RFA-LCF integrates the robust flexible CF by clean data space recovery, robust sparse local-coordinate coding, and adaptive weighting into a unified model. RFA-LCF improves the representations by enhancing the robustness of CF to noise and errors, providing a flexible constraint on the reconstruction error and optimizing the locality jointly. For robust learning, RFA-LCF clearly learns a sparse projection to recover the underlying clean data space, and then the flexible CF is performed in the projected feature space. RFA-LCF also uses a L2,1-norm based flexible residue to encode the mismatch between the recovered data and its reconstruction, and uses the robust sparse local-coordinate coding to represent data using a few nearby basis concepts. For auto-weighting, RFA-LCF jointly preserves the manifold structures in the basis concept space and new coordinate space in an adaptive manner by minimizing the reconstruction errors on clean data, anchor points and coordinates. By updating the local-coordinate preserving data, basis concepts and new coordinates alternately, the representation abilities can be potentially improved. Extensive results on public databases show that RFA-LCF delivers the state-of-the-art clustering results compared with other related methods.

With the popularity of multimedia technology, information is always represented from multiple views. Even though multiview data can reflect the same sample from different perspectives, multiple views are consistent to some extent because they are representations of the same sample. Most of the existing algorithms are graph-based ones to learn the complex structures within multiview data but overlook the information within data representations. Furthermore, many existing works treat multiple views discriminatively by introducing some hyperparameters, which is undesirable in practice. To this end, abundant multiview-based methods have been proposed for dimension reduction. However, there is still no research that leverages the existing work into a unified framework. In this paper, we propose a general framework for multiview data dimension reduction, named kernelized multiview subspace analysis (KMSA) to handle multiview feature representation in the kernel space, providing a feasible channel for multiview data with different dimensions. Compared with the graph-based methods, KMSA can fully exploit information from multiview data with nothing to lose. Since different views have different influences on KMSA, we propose a self-weighted strategy to treat different views discriminatively. A co-regularized term is proposed to promote the mutual learning from multiviews. KMSA combines self-weighted learning with the co-regularized term to learn the appropriate weights for all views. We evaluate our proposed framework on 6 multiview datasets for classification and image retrieval. The experimental results validate the advantages of our proposed method.

The projective dictionary pair learning (DPL) model jointly seeks a synthesis dictionary and an analysis dictionary by extracting the block-diagonal coefficients with an incoherence-constrained analysis dictionary. However, DPL fails to discover the underlying subspaces and salient features at the same time, and it cannot encode the neighborhood information of the embedded coding coefficients, especially adaptively. In addition, although the data can be well reconstructed via the minimization of the reconstruction error, useful distinguishing salient feature information may be lost and incorporated into the noise term. In this article, we propose a novel self-expressive adaptive locality-preserving framework: twin-incoherent self-expressive latent DPL (SLatDPL). To capture the salient features from the samples, SLatDPL minimizes a latent reconstruction error by integrating the coefficient learning and salient feature extraction into a unified model, which can also be used to simultaneously discover the underlying subspaces and salient features. To make the coefficients block diagonal and ensure that the salient features are discriminative, our SLatDPL regularizes them by imposing a twin-incoherence constraint. Moreover, SLatDPL utilizes a self-expressive adaptive weighting strategy that uses normalized block-diagonal coefficients to preserve the locality of the codes and salient features. SLatDPL can use the class-specific reconstruction residual to handle new data directly. Extensive simulations on several public databases demonstrate the satisfactory performance of our SLatDPL compared with related methods.

Multioutput regression, referring to simultaneously predicting multiple continuous output variables with a single model, has drawn increasing attention in the machine learning community due to its strong ability to capture the correlations among multioutput variables. The methodology of output space embedding, built upon the low-rank assumption, is now the mainstream for multioutput regression since it can effectively reduce the parameter numbers while achieving effective performance. The existing low-rank methods, however, are sensitive to the noises of both inputs and outputs, referring to the noise problem. In this article, we develop a novel multioutput regression method by simultaneously alleviating input and output noises, namely, robust multioutput regression by alleviating input and output noises (RMoR-Aion), where both the noises of the input and output are exploited by leveraging auxiliary matrices. Furthermore, we propose a prediction output manifold constraint with the correlation information regarding the output variables to further reduce the adversarial effects of the noise. Our empirical studies demonstrate the effectiveness of RMoR-Aion compared with the state-of-the-art baseline methods, and RMoR-Aion is more stable in the settings with artificial noise.

Many studies on automated COVID-19 diagnosis have advanced rapidly with the increasing availability of large-scale CT annotated datasets. Inevitably, there are still a large number of unlabeled CT slices in the existing data sources since it requires considerable consuming labor efforts. Notably, cinical experience indicates that the neighboring CT slices may present similar symptoms and signs. Inspired by such wisdom, we propose DACE, a novel CNN-based deep active context estimation framework, which leverages the unlabeled neighbors to progressively learn more robust feature representations and generate a well-performed classifier for COVID-19 diagnosis. Specifically, the backbone of the proposed DACE framework is constructed by a well-designed Long-Short Hierarchical Attention Network (LSHAN), which effectively incorporates two complementary attention mechanisms, i.e., short-range channel interactions (SCI) module and long-range spatial dependencies (LSD) module, to learn the most discriminative features from CT slices. To make full use of such available data, we design an efficient context estimation criterion to carefully assign the additional labels to these neighbors. Benefiting from two complementary types of informative annotations from -nearest neighbors, i.e., the majority of high-confidence samples with pseudo labels and the minority of low-confidence samples with hand-annotated labels, the proposed LSHAN can be fine-tuned and optimized in an incremental learning manner. Extensive experiments on the Clean-CC-CCII dataset demonstrate the superior performance of our method compared with the state-of-the-art baselines.

Low-rank coding-based representation learning is powerful for discovering and recovering the subspace structures in data, which has obtained an impressive performance; however, it still cannot obtain deep hidden information due to the essence of single-layer structures. In this article, we investigate the deep low-rank representation of images in a progressive way by presenting a novel strategy that can extend existing single-layer latent low-rank models into multiple layers. Technically, we propose a new progressive Deep Latent Low-Rank Fusion Network (DLRF-Net) to uncover deep features and the clustering structures embedded in latent subspaces. The basic idea of DLRF-Net is to progressively refine the principal and salient features in each layer from previous layers by fusing the clustering and projective subspaces, respectively, which can potentially learn more accurate features and subspaces. To obtain deep hidden information, DLRF-Net inputs shallow features from the last layer into subsequent layers. Then, it aims at recovering the hierarchical information and deeper features by respectively congregating the subspaces in each layer of the network. As such, one can also ensure the representation learning of deeper layers to remove the noise and discover the underlying clean subspaces, which will be verified by simulations. It is noteworthy that the framework of our DLRF-Net is general and is applicable to most existing latent low-rank representation models, i.e., existing latent low-rank models can be easily extended to the multilayer scenario using DLRF-Net. Extensive results on real databases show that our framework can deliver enhanced performance over other related techniques.

For subspace recovery, most existing low-rank representation (LRR) models performs in the original space in single-layer mode. As such, the deep hierarchical information cannot be learned, which may result in inaccurate recoveries for complex real data. In this paper, we explore the deep multi-subspace recovery problem by designing a multilayer architecture for latent LRR. Technically, we propose a new Multilayer Collaborative Low-Rank Representation Network model termed DeepLRR to discover deep features and deep subspaces. In each layer (>2), DeepLRR bilinearly reconstructs the data matrix by the collaborative representation with low-rank coefficients and projection matrices in the previous layer. The bilinear low-rank reconstruction of previous layer is directly fed into the next layer as the input and low-rank dictionary for representation learning, and is further decomposed into a deep principal feature part, a deep salient feature part and a deep sparse error. As such, the coherence issue can be also resolved due to the low-rank dictionary, and the robustness against noise can also be enhanced in the feature subspace. To recover the sparse errors in layers accurately, a dynamic growing strategy is used, as the noise level will become smaller for the increase of layers. Besides, a neighborhood reconstruction error is also included to encode the locality of deep salient features by deep coefficients adaptively in each layer. Extensive results on public databases show that our DeepLRR outperforms other related models for subspace discovery and clustering.

The Deep Convolutional Neural Networks (CNNs) have obtained a great success for pattern recognition, such as recognizing the texts in images. But existing CNNs based frameworks still have several drawbacks: 1) the traditaional pooling operation may lose important feature information and is unlearnable; 2) the tradi-tional convolution operation optimizes slowly and the hierar-chical features from different layers are not fully utilized. In this work, we address these problems by developing a novel deep network model called Fully-Convolutional Intensive Feature Flow Neural Network (IntensiveNet). Specifically, we design a further dense block called intensive block to extract the feature information, where the original inputs and two dense blocks are connected tightly. To encode data appropriately, we present the concepts of dense fusion block and further dense fusion opera-tions for our new intensive block. By adding short connections to different layers, the feature flow and coupling between layers are enhanced. We also replace the traditional convolution by depthwise separable convolution to make the operation efficient. To prevent important feature information being lost to a certain extent, we use a convolution operation with stride 2 to replace the original pooling operation in the customary transition layers. The recognition results on large-scale Chinese string and MNIST datasets show that our IntensiveNet can deliver enhanced recog-nition results, compared with other related deep models.

Dictionary learning (DL) is powerful for representation learning, while it fails to capture the deep hierarchical information hidden in data. In this paper, we propose a new generalized end-to-end mulita-layer representation learning architecture referred to as Multi-layer Dictionary Pair Learning Network (MDPL-net) for the deep sparse and hierarchical representation of images. To enable MDPL-net to conduct accurate classification, MDPL-net clearly integrates the skip connection end-to-end network and multi-layer deep sparse dictionary learning into a unified architecture. The representation learning module has several hidden DL blocks, where each hidden DL block has a dictionary pair learning (DPL) layer, a batch-norm layer and an activation function layer, and the DL blocks are connected in a feed-forward manner. To further improve the information flow and maintain the privileged features between different DL blocks, a novel skip dense connectivity pattern is deployed between hidden DL blocks, which can obtain more stable and discriminative features. The DPL layer jointly formulates the discriminative synthesis dictionary and analysis dictionary by minimizing reconstruction error within each batch over the feature maps from front layers. Extensive results on benchmark databases demonstrate the effectiveness of MDPL-net for discriminative representation and robust image classification.

Vehicle re-identification (reID) aims at identifying vehicles across different non-overlapping cameras views. The existing methods heavily relied on well-labeled datasets for ideal performance, which inevitably causes fateful drop due to the severe domain bias between the training domain and the real-world scenes; worse still, these approaches required full annotations, which is labor-consuming. To tackle these challenges, we propose a novel Progressive Adaptation Learning method for vehicle reID, named PAL, which infers from the abundant data without annotations. For PAL, a data adaptation module is employed for source domain, which generates the images with similar data distribution to unlabeled target domain as “pseudo target samples”. These pseudo samples are combined with the unlabeled samples that are selected by a dynamic sampling strategy to make training faster. We further proposed a weighted label smoothing (WLS) loss, which considers the similarity between samples with different clusters to balance the confidence of pseudo labels. Comprehensive experimental results validate the advantages of PAL on both VehicleID and VeRi-776 dataset.

Low-rank representation is powerful for recover-ing and clustering the subspace structures, but it cannot obtain deep hierarchical information due to the single-layer mode. In this paper, we present a new and effective strategy to extend the sin-gle-layer latent low-rank models into multi-ple-layers, and propose a new and progressive Deep Latent Low-Rank Fusion Network (DLRF-Net) to uncover deep features and struc-tures embedded in input data. The basic idea of DLRF-Net is to refine features progressively from the previous layers by fusing the subspaces in each layer, which can potentially obtain accurate fea-tures and subspaces for representation. To learn deep information, DLRF-Net inputs shallow fea-tures of the last layers into subsequent layers. Then, it recovers the deeper features and hierar-chical information by congregating the projective subspaces and clustering subspaces respectively in each layer. Thus, one can learn hierarchical sub-spaces, remove noise and discover the underlying clean subspaces. Note that most existing latent low-rank coding models can be extended to multi-layers using DLRF-Net. Extensive results show that our network can deliver enhanced perfor-mance over other related frameworks.

In this paper, a new deep incomplete multi-view clustering network, called DIMC-net, is proposed to address the challenge of multi-view clustering on missing views. In particular, DIMC-net designs several view-specific encoders to extract the high-level information of multiple views and introduces a fusion graph based constraint to explore the local geometric information of data. To reduce the negative influence of missing views, a weighted fusion layer is introduced to obtain the consensus representation shared by all views. Moreover, a clustering layer is introduced to guarantee that the obtained consensus representation is the best one for the clustering task. Compared with the existing deep learning based approaches, DIMC-net is more flexible and efficient since it can handle all kinds of incomplete cases and directly produce the clustering results. Experimental results show that DIMC-net achieves significant improvement over state-of-the-art incomplete multi-view clustering methods.

In this paper, we technically propose a novel framework called Deep Self-representative Concept Factorization Network (DSCF-Net), for clustering deep features. To improve the representation and clustering abilities, DSCF-Net explicitly considers discovering hidden deep semantic features, enhancing the robustness properties of the deep factorization to noise and preserving the local manifold structures of deep features. Specifically, DSCF-Net integrates the robust deep concept factorization, deep self-expressive representation and adaptive locality preserving feature learning into a unified framework. To discover hidden deep representations, DSCF-Net designs a hierarchical factorization architecture using multiple layers of linear transformations, where the hierarchical representation is performed by formulating the problem as optimizing the basis concepts in each layer to improve the representation indirectly. DSCF-Net also improves robustness by subspace recovery for sparse error correction firstly and then performs deep factorization in the recovered visual subspace. To obtain localitypreserving representations, we also present an adaptive deep self-representative weighting strategy by using the coefficient matrix as adaptive weights to keep the locality of representations. Extensive results show that DSCF-Net delivers state-of-the-art performance on several public databases.

We propose a joint subspace recovery and enhanced locality-based robust flexible label consistent dictionary learning method called Robust Flexible Discriminative Dictionary Learning (RFDDL). The RFDDL mainly improves the data representation and classification abilities by enhancing the robust property to sparse errors and encoding the locality, reconstruction error, and label consistency more accurately. First, for the robustness to noise and sparse errors in data and atoms, the RFDDL aims at recovering the underlying clean data and clean atom subspaces jointly, and then performs DL and encodes the locality in the recovered subspaces. Second, to enable the data sampled from a nonlinear manifold to be handled potentially and obtain the accurate reconstruction by avoiding the overfitting, the RFDDL minimizes the reconstruction error in a flexible manner. Third, to encode the label consistency accurately, the RFDDL involves a discriminative flexible sparse code error to encourage the coefficients to be soft. Fourth, to encode the locality well, the RFDDL defines the Laplacian matrix over recovered atoms, includes label information of atoms in terms of intra-class compactness and inter-class separation, and associates with group sparse codes and classifier to obtain the accurate discriminative locality-constrained coefficients and classifier. The extensive results on public databases show the effectiveness of our RFDDL.

Zero-shot learning (ZSL) is a challenging task due to the lack of unseen class data during training. Existing works attempt to establish a mapping between the visual and class spaces through a common intermediate semantic space. The main limitation of existing methods is the strong bias towards seen class, known as the domain shift problem, which leads to unsatisfactory performance in both conventional and generalized ZSL tasks. To tackle this challenge, we propose to convert ZSL to the conventional supervised learning by generating features for unseen classes. To this end, a joint generative model that couples variational autoencoder (VAE) and generative adversarial network (GAN), called Zero-VAE-GAN, is proposed to generate high-quality unseen features. To enhance the class-level discriminability, an adversarial categorization network is incorporated into the joint framework. Besides, we propose two self-training strategies to augment unlabeled unseen features for the transductive extension of our model, addressing the domain shift problem to a large extent. Experimental results on five standard benchmarks and a large-scale dataset demonstrate the superiority of our generative model over the state-of-the-art methods for conventional, especially generalized ZSL tasks. Moreover, the further improvement of the transductive setting demonstrates the effectiveness of the proposed self-training strategies.

In this paper, we investigate the robust dictionary learning (DL) to discover the hybrid salient low-rank and sparse representation in a factorized compressed space. A Joint Robust Factorization and Projective Dictionary Learning (J-RFDL) model is presented. The setting of J-RFDL aims at improving the data representations by enhancing the robustness to outliers and noise in data, encoding the reconstruction error more accurately and obtaining hybrid salient coefficients with accurate reconstruction ability. Specifically, J-RFDL performs the robust representation by DL in a factorized compressed space to eliminate the negative effects of noise and outliers on the results, which can also make the DL process efficient. To make the encoding process robust to noise in data, J-RFDL clearly uses sparse L2, 1-norm that can potentially minimize the factorization and reconstruction errors jointly by forcing rows of the reconstruction errors to be zeros. To deliver salient coefficients with good structures to reconstruct given data well, J-RFDL imposes the joint low-rank and sparse constraints on the embedded coefficients with a synthesis dictionary. Based on the hybrid salient coefficients, we also extend J-RFDL for the joint classification and propose a discriminative J-RFDL model, which can improve the discriminating abilities of learnt coefficients by minimizing the classification error jointly. Extensive experiments on public datasets demonstrate that our formulations can deliver superior performance over other state-of-the-art methods.

Constrained Concept Factorization (CCF) yields the enhanced representation ability over CF by incorporating label information as additional constraints, but it cannot classify and group unlabeled data appropriately. Minimizing the difference between the original data and its reconstruction directly can enable CCF to model a small noisy perturbation, but is not robust to gross sparse errors. Besides, CCF cannot preserve the manifold structures in new representation space explicitly, especially in an adaptive manner. In this paper, we propose a joint label prediction based Robust Semi-Supervised Adaptive Concept Factorization (RS2ACF) framework. To obtain robust representation, RS2ACF relaxes the factorization to make it simultaneously stable to small entrywise noise and robust to sparse errors. To enrich prior knowledge to enhance the discrimination, RS2ACF clearly uses class information of labeled data and more importantly propagates it to unlabeled data by jointly learning an explicit label indicator for unlabeled data. By the label indicator, RS2ACF can ensure the unlabeled data of the same predicted label to be mapped into the same class in feature space. Besides, RS2ACF incorporates the joint neighborhood reconstruction error over the new representations and predicted labels of both labeled and unlabeled data, so the manifold structures can be preserved explicitly and adaptively in the representation space and label space at the same time. Owing to the adaptive manner, the tricky process of determining the neighborhood size or kernel width can be avoided. Extensive results on public databases verify that our RS2ACF can deliver state-of-the-art data representation, compared with other related methods.

Both interclass variances and intraclass similarities are crucial for improving the classification performance of discriminative dictionary learning (DDL) algorithms. However, existing DDL methods often ignore the combination between the interclass and intraclass properties of dictionary atoms and coding coefficients. To address this problem, in this paper, we propose a discriminative Fisher embedding dictionary learning (DFEDL) algorithm that simultaneously establishes Fisher embedding models on learned atoms and coefficients. Specifically, we first construct a discriminative Fisher atom embedding model by exploring the Fisher criterion of the atoms, which encourages the atoms of the same class to reconstruct the corresponding training samples as much as possible. At the same time, a discriminative Fisher coefficient embedding model is formulated by imposing the Fisher criterion on the profiles (row vectors of the coding coefficient matrix) and coding coefficients, which forces the coding coefficient matrix to become a block-diagonal matrix. Since the profiles can indicate which training samples are represented by the corresponding atoms, the proposed two discriminative Fisher embedding models can alternatively and interactively promote the discriminative capabilities of the learned dictionary and coding coefficients. The extensive experimental results demonstrate that the proposed DFEDL algorithm achieves superior performance in comparison with some state-of-the-art dictionary learning algorithms on both hand-crafted and deep learning-based features.

The choice of which clothes to wear affects how one is perceived, as well as constitutes an expression of one's personal style. Based on the recent advances in image-to-image translation by the conditional generative adversarial network (cGAN), we propose a new framework with a multidiscriminator by incorporating different types of conditional information into the discriminator of cGAN for clothing matches. In contrast with most extant frameworks under cGAN, with one generator and one discriminator, the proposed framework investigates the potential of utilizing conditional information delivered by multidiscriminators to guide the generator. Under this framework, we propose an Attribute-GAN with two discriminators and a category-attribute GAN (CA-GAN) with three discriminators. In order to evaluate the performance of our proposed models, we built a large-scale data set that consists of 19 081 pairs of collocation clothing images with 90 manually labeled attributes. Experimental results demonstrate that with supervision of the additional attribute discriminator or category discriminator, the quality of the generated clothing images by GANs is consistently improved in comparison with the state-of-the-art methods.

In this article, we propose a structured robust adaptive dictionary pair learning (RA-DPL) framework for the discriminative sparse representation (SR) learning. To achieve powerful representation ability of the available samples, the setting of RA-DPL seamlessly integrates the robust projective DPL, locality-adaptive SRs, and discriminative coding coefficients learning into a unified learning framework. Specifically, RA-DPL improves existing projective DPL in four perspectives. First, it applies a sparse l2,1-norm-based metric to encode the reconstruction error to deliver the robust projective dictionary pairs, and the l2,1-norm has the potential to minimize the error. Second, it imposes the robust l2,1-norm clearly on the analysis dictionary to ensure the sparse property of the coding coefficients rather than using the costly l0/l1-norm. As such, the robustness of the data representation and the efficiency of the learning process are jointly considered to guarantee the efficacy of our RA-DPL. Third, RA-DPL conceives a structured reconstruction weight learning paradigm to preserve the local structures of the coding coefficients within each class clearly in an adaptive manner, which encourages to produce the locality preserving representations. Fourth, it also considers improving the discriminating ability of coding coefficients and dictionary by incorporating a discriminating function, which can ensure high intraclass compactness and interclass separation in the code space. Extensive experiments show that our RA-DPL can obtain superior performance over other state of the arts.

The graph-based semisupervised label propagation (LP) algorithm has delivered impressive classification results. However, the estimated soft labels typically contain mixed signs and noise, which cause inaccurate predictions due to the lack of suitable constraints. Moreover, the available methods typically calculate the weights and estimate the labels in the original input space, which typically contains noise and corruption. Thus, the encoded similarities and manifold smoothness may be inaccurate for label estimation. In this article, we present effective schemes for resolving these issues and propose a novel and robust semisupervised classification algorithm, namely the triple matrix recovery-based robust auto-weighted label propagation framework (ALP-TMR). Our ALP-TMR introduces a TMR mechanism to remove noise or mixed signs from the estimated soft labels and improve the robustness to noise and outliers in the steps of assigning weights and predicting the labels simultaneously. Our method can jointly recover the underlying clean data, clean labels, and clean weighting spaces by decomposing the original data, predicted soft labels, or weights into a clean part plus an error part by fitting noise. In addition, ALP-TMR integrates the auto-weighting process by minimizing the reconstruction errors over the recovered clean data and clean soft labels, which can encode the weights more accurately to improve both data representation and classification. By classifying samples in the recovered clean label and weight spaces, one can potentially improve the label prediction results. Extensive simulations verified the effectivenss of our ALP-TMR.

We investigate the high-dimensional data clustering problem by proposing a novel and unsupervised representation learning model called Robust Flexible Auto-weighted Local-coordinate Concept Factorization (RFA-LCF). RFA-LCF integrates the robust flexible CF, robust sparse local-coordinate coding and the adaptive reconstruction weighting learning into a unified model. The adaptive weighting is driven by including thejoint manifold preserving constraints on the recovered clean data, basis concepts and new representation. Specifically, our RFA-LCF uses a L2,1-norm based flexible residue to encode the mismatch between clean data and its reconstruction, and also applies the robust adaptive sparse local-coordinate coding to represent the data using a few nearby basis concepts, which can make the factorization more accurate and robust to noise. The robust flexible factorization is also performed in the recovered clean data space for enhancing representations. RFA-LCF also considers preserving the local manifold structures of clean data space, basis concept space and the new coordinate space jointly in an adaptive manner way. Extensive comparisons show that RFA-LCF can deliver enhanced clustering results.

In this paper, we propose a robust representation learning model called Adaptive Structure-constrained Low-Rank Coding (AS-LRC) for the latent representation of data. To recover the underlying subspaces more accurately, AS-LRC seamlessly integrates an adaptive weighting based block-diagonal structure-constrained low-rank representation and the group sparse salient feature extraction into a unified framework. Specifically, AS-LRC performs the latent decomposition of given data into a low-rank reconstruction by a block-diagonal codes matrix, a group sparse locality-adaptive salient feature part and a sparse error part. To enforce the block-diagonal structures adaptive to different real datasets for the low-rank recovery, AS-LRC clearly computes an auto-weighting matrix based on the locality-adaptive features and multiplies by the low-rank coefficients for direct minimization at the same time. This encourages the codes to be block-diagonal and can avoid the tricky issue of choosing optimal neighborhood size or kernel width for the weight assignment, suffered in most local geometrical structures-preserving low-rank coding methods. In addition, our AS-LRC selects the L2, 1-norm on the projection for extracting group sparse features rather than learning low-rank features by Nuclear-norm regularization, which can make learnt features robust to noise and outliers in samples, and can also make the feature coding process efficient. Extensive visualizations and numerical results demonstrate the effectiveness of our AS-LRC for image representation and recovery.

Single image deraining task is still a very challenging task due to its ill-posed nature in reality. Recently, researchers have tried to fix this issue by training the CNN-based end-to-end models, but they still cannot extract the negative rain streaks from rainy images precisely, which usually leads to an over de-rained or under de-rained result. To handle this issue, this paper proposes a new coarse-to-fine single image deraining framework termed Multi-stream Hybrid Deraining Network (shortly, MH-DerainNet). To obtain the negative rain streaks during training process more accurately, we present a new module named dual path residual dense block, i.e., Residual path and Dense path. The Residual path is used to reuse com-mon features from the previous layers while the Dense path can explore new features. In addition, to concatenate different scaled features, we also apply the idea of multi-stream with shortcuts between cascaded dual path residual dense block based streams. To obtain more distinct derained images, we combine the SSIM loss and perceptual loss to preserve the per-pixel similarity as well as preserving the global structures so that the deraining result is more accurate. Extensive experi-ments on both synthetic and real rainy images demonstrate that our MH-DerainNet can deliver significant improvements over several recent state-of-the-art methods.

In this paper, we extend the popular dictionary pair learning (DPL) into the scenario of twin-projective latent flexible DPL under a structured twin-incoherence. Technically, a novel framework called Twin-Projective Latent Flexible DPL (TP-DPL) is proposed, which minimizes the twin-incoherence constrained flexibly-relaxed reconstruction error to avoid the possible over-fitting issue and produce accurate reconstruction. In this setting, TP-DPL integrates the twin-incoherence based latent flexible DPL and the joint embedding of codes as well as salient features by twin-projection into a unified model in an adaptive neighborhood-preserving manner. Therefore, TP-DPL can unify the procedures of salient feature representation and classification. The twin-incoherence constraint on coefficients and features can explicitly ensure high intra-class compactness and inter-class separation over them. TP-DPL also integrates the adaptive weighting to preserve local neighborhood of both coefficients and salient features within each class explicitly. For efficiency, TP-DPL selects the Frobenius-norm and abandons the costly l0/l1-norm for group sparse representation. Another byproduct is that TP-DPL can directly apply the class-specific twin-projective reconstruction residual to compute the label of data. Extensive results on public databases show that TP-DPL can deliver the state-of-the-art performance.

We propose a novel structured discriminative block- diagonal dictionary learning method, referred to as scalable Locality-Constrained Projective Dictionary Learning (LC-PDL), for efficient representation and classification. To improve the scalability by saving both training and testing time, our LC-PDL aims at learning a structured discriminative dictionary and a block-diagonal representation without using costly l0/l1-norm. Besides, it avoids extra time-consuming sparse reconstruction process with the well-trained dictionary for new sample as many existing models. More importantly, LC-PDL avoids using the com- plementary data matrix to learn the sub-dictionary over each class. To enhance the performance, we incorporate a locality constraint of atoms into the DL procedures to keep local information and obtain the codes of samples over each class separately. A block-diagonal discriminative approximation term is also derived to learn a discriminative projection to bridge data with their codes by extracting the special block-diagonal features from data, which can ensure the approximate coefficients to associate with its label information clearly. Then, a robust multiclass classifier is trained over extracted block-diagonal codes for accurate label predictions. Experimental results verify the effectiveness of our algorithm.

We propose a novel and unsupervised representation learning model, i.e., Robust Block-Diagonal Adaptive Locality-constrained Latent Representation (rBDLR). rBDLR is able to recover multi-subspace structures and extract the adaptive locality-preserving salient features jointly. Leveraging on the Frobenius-norm based latent low-rank representation model, rBDLR jointly learns the coding coefficients and salient features, and improves the results by enhancing the robustness to outliers and errors in given data, preserving local information of salient features adaptively and ensuring the block-diagonal structures of the coefficients. To improve the robustness, we perform the latent representation and adaptive weighting in a recovered clean data space. To force the coefficients to be block-diagonal, we perform auto-weighting by minimizing the reconstruction error based on salient features, constrained using a block-diagonal regularizer. This ensures that a strict block-diagonal weight matrix can be obtained and salient features will possess the adaptive locality preserving ability. By minimizing the difference between the coefficient and weights matrices, we can obtain a block-diagonal coefficients matrix and it can also propagate and exchange useful information between salient features and coefficients. Extensive results demonstrate the superiority of rBDLR over other state-of-the-art methods.

Kernel methods have been successfully applied to the areas of pattern recognition and data mining. In this paper, we mainly discuss the issue of propagating labels in kernel space. A Kernel-Induced Label Propagation (Kernel-LP) framework by mapping is proposed for high-dimensional data classification using the most informative patterns of data in kernel space. The essence of Kernel-LP is to perform joint label propagation and adaptive weight learning in a transformed kernel space. That is, our Kernel-LP changes the task of label propagation from the commonly-used Euclidean space in most existing work to kernel space. The motivation of our Kernel-LP to propagate labels and learn the adaptive weights jointly by the assumption of an inner product space of inputs, i.e., the original linearly inseparable inputs may be mapped to be separable in kernel space. Kernel-LP is based on existing positive and negative LP model, i.e., the effects of negative label information are integrated to improve the label prediction power. Also, Kernel-LP performs adaptive weight construction over the same kernel space, so it can avoid the tricky process of choosing the optimal neighborhood size suffered in traditional criteria. Two novel and efficient out-of-sample approaches for our Kernel-LP to involve new test data are also presented, i.e., (1) direct kernel mapping and (2) kernel mapping-induced label reconstruction, both of which purely depend on the kernel matrix between training set and testing set. Owing to the kernel trick, our algorithms will be applicable to handle the high-dimensional real data. Extensive results on real datasets demonstrate the effectiveness of our approach.

Dimension reduction is widely regarded as an effective way for decreasing the computation, storage and communication loads of data-driven intelligent systems, leading to a growing demand for statistical methods that allow analysis (e.g., clustering) of compressed data. We therefore study in this paper a novel problem called compressive robust subspace clustering, which is to perform robust subspace clustering with the compressed data, and which is generated by projecting the original high-dimensional data onto a lower-dimensional subspace chosen at random. Given only the compressed data and sensing matrix, the proposed method, row space pursuit (RSP), recovers the authentic row space that gives correct clustering results under certain conditions. Extensive experiments show that RSP is distinctly better than the competing methods, in terms of both clustering accuracy and computational efficiency.

In this paper, we propose a novel unsupervised Nonnegative Adaptive Feature Extraction (NAFE) algorithm for data representation and classification. The formulation of NAFE integrates the sparsity constrained nonnegative matrix factorization (NMF), representation learning, and adaptive reconstruction weight learning into a unified model. Specifically, NAFE performs feature and weight learning over the new robust representations of NMF for more accurate measure and representation. For nonnegative adaptive feature extraction, our NAFE first utilizes the sparsity constrained NMF to obtain the new and robust representations of the original data. To preserve the manifold structures of the learnt new representations, we also incorporate a neighborhood reconstruction error over the weight matrix for joint minimization. Note that to further improve the representation power, the weights are jointly shared in the new low-dimensional nonnegative representation space, low-dimensional nonlinear manifold space, and low-dimensional projective subspace, i.e., local neighborhood information is clearly preserved in different feature spaces so that informative representations and features can be jointly obtained. To enable NAFE to extract features from new data, we also include a feature approximation error by a linear projection so that the learnt extractor can obtain features from new data efficiently. Extensive simulations show that our formulation can deliver state-of-the-art results on several public databases for feature extraction and classification, compared with several related methods.

Of late, there are many studies on the robust discriminant analysis, which adopt L1-norm as the distance metric, but their results are not robust enough to gain universal acceptance. To overcome this problem, the authors of this article present a nonpeaked discriminant analysis (NPDA) technique, in which cutting L1-norm is adopted as the distance metric. As this kind of norm can better eliminate heavy outliers in learning models, the proposed algorithm is expected to be stronger in performing feature extraction tasks for data representation than the existing robust discriminant analysis techniques, which are based on the L1-norm distance metric. The authors also present a comprehensive analysis to show that cutting L1-norm distance can be computed equally well, using the difference between two special convex functions. Against this background, an efficient iterative algorithm is designed for the optimization of the proposed objective. Theoretical proofs on the convergence of the algorithm are also presented. Theoretical insights and effectiveness of the proposed method are validated by experimental tests on several real data sets.

Named Data Networking (NDN), a revolution of the IP architecture, provides an information-centric paradigm to transmit video traffic which occupies most of the Internet traffic. However, the control schemes in the field of video NDN research can easily result in full bottleneck bandwidth at the cost of high video delay and frequent packet loss. This is because the loss-based or transient-queue-delay-based congestion control schemes respond to the signal after the congestion is formed. And NDN multi-source feature can't make the loss and delay-based signals accurately reflect the network situation. In addition, the current video applications need to choose the suitable video rate to stream across the network. The bandwidth estimation scheme of video rate control ignoring multi-source reduces accuracy in NDN. In this paper, we firstly build a max-plus algebra based model to characterize the interest-data transmission process for video consumer and to estimate the NDN bandwidth delay product. The hops (lived time)that data pass through from the producer to the consumer represent the multi-source feature in the model. Then we propose a Time-Lived based congestion and rate control scheme for the NDN video consumer. Extensive experiments have been conducted to evaluate our scheme from the perspective of throughput, delay, and queue length. The results show that our scheme can improve the throughput of the video consumer while reducing the round trip time by shortening the queue length.

In this paper, we discuss the robust factorization based robust dictionary learning problem for data representation. A Robust Projective Low-Rank and Sparse Representation model (R-PLSR) is technically proposed. Our R-PLSR model integrates the L1-norm based robust factorization and robust low-rank & sparse representation by robust dictionary learning into a unified framework. Specifically, R-PLSR performs the joint low-rank and sparse representation over the informative low-dimensional representations by robust sparse factorization so that the results are more accurate. To make the factorization and representation procedures robust to noise and outliers, R-PLSR imposes the sparse L2, 1-norm jointly on the reconstruction errors based on the factorization and dictionary learning. Note that L2, 1-norm can also minimize the reconstruction error as much as possible, since the L2, 1-norm theoretically tends to force many rows of the reconstruction error matrix to be zeros. The Nuclear-norm and L1-norm are jointly used on the representation coefficients so that salient representations can be obtained. Extensive results on several image datasets show that our R-PLSR formulation can deliver superior performance over other state-of-the-arts.

In this paper, we mainly propose a Robust Adaptive Projective Dictionary Pair Learning (RA-DPL) framework based on the adaptive discriminative representations. Our formulation can seamlessly integrate the robust projective dictionary pair learning and the adaptive sparse representation learning into a unified model. RA-DPL improves the existing DPL algorithm in threefold. First, RA-DPL aims at computing the robust projective dictionary pairs by employing the sparse and robust l2,1-norm to encode the reconstruction error. Second, RA-DPL regularizes the robust l2,1-norm on the analysis dictionary so that the analysis dictionary can extract sparse coefficients from the given samples explicitly. More importantly, the optimization of l2,1-norm is so efficient, that is, the sparse coding step will be time-saving. Third, RA-DPL can clearly preserve the local neighborhood relationship of the sparse coefficients within each class, which can make the learnt representations discriminating and can also improve the discriminating power of learnt dictionary. Extensive simulations on image databases demonstrate that our RA-DPL can obtain the superior performance over other state-of-the-arts.

Most existing low-rank sparse embedding models extract features of data in the original input space and usually separate the manifold preservation step from the coding process, which may result in the decreased performance. In this paper, a novel Robust Adaptive Low-rank and Sparse Embedding (RALSE) framework is technically proposed for salient feature extraction of the high-dimensional data by seamlessly integrating the joint low-rank and sparse recovery with the robust adaptive salient feature extraction. Specifically, our RALSE integrates the joint low-rank and sparse representation, adaptive neighborhood preserving graph weight learning and the robustness-promoting representation into a unified framework. For accurate similarity measure, RALSE computes the adaptive weights by minimizing the reconstruction error over the noise-removed data and salient features simultaneously, where L1-norm is regularized to ensure the sparse properties of learnt weights. RALSE can also ensure the learnt projection to preserve local neighborhood information of embedded features clearly and adaptively. The projection is not only modeled under joint low-rank and sparse regularization, but also computed from a clean subspace, making it powerful for the salient feature extraction. Thus, the learnt low-rank sparse features would be more accurate for subsequent classification. Extensive results demonstrate the effectiveness of our RALSE formulation for data representation and classification.

A novel discriminant analysis technique for feature extraction, referred to as Robust Discriminant Subspace (RDS) with L2,p+s-Norm Distance Maximization-Minimization (maxmin) is posed. In its objective, the within-class and between-class distances are measured by L2,p-norm and L2,s-norm, respectively, such that it is robust and rotational invariant. An efficient iterative algorithm is designed to solve the resulted objective, which is non-greedy. We also conduct some insightful analysis on the convergence of the proposed algorithm. Theoretical insights and effectiveness of our RDS are further supported by promising experimental results on several images databases.

We mainly propose a robust Embedded Locality-Constrained Label Consistent Dictionary Learning (ELC2DL) framework for discriminative classification. ELC2DL improves the representation and classification performance by performing DL in the noise-removed sparse embedding space, since most real data often contains noise and performing DL over noisy data for reconstruction may decrease performance potentially. To reduce the noise in data, our model computes a sparse projection jointly for noise reduction and then uses the noise-removed data for DL. By incorporating a noise-reduction term with a discriminative locality-constrained label consistent term that associates the label information with each dictionary atom to preserve local structure of training data, a noise-reduction projection, an over-complete dictionary and discriminative sparse codes are obtained jointly. Simulations on several image databases show that our algorithm can deliver enhanced performance over other state-of-the-arts.

In this paper, we investigate the robust transductive label prediction problem. Technically, a Robust Adaptive Label Propagation framework by Double Matrix Decomposition, called ALP-MD, is proposed for the semi-supervised data classification. Compared with existing transductive label propagation models, our ALP-MD improves the classification power by performing label prediction in the clean data space and clean label space at the same time. More specifically, our ALP-MD clearly integrates the idea of double matrix decomposition into the process of label prediction for the noise removal. Since the predicted soft labels usually contains noise and mixed signs, our ALP-MD explicitly decomposes the predicted soft label matrix into a clean soft label matrix and a noise term and then estimates the hard label based on the clean soft label matrix for more accurate classification. In addition, ALP-MD also involves a regularization term to model the noise in data, integrates the adaptive weights learning into the process of robust label prediction and moreover performs the weights learning in the clean data space. Thus, our ALP-MD can explicitly ensure the learned weights to be informative as much as possible and to be joint optimal for both representation and classification, and potentially enhance the label prediction ability. Extensive comparisons demonstrated its effectiveness.

In this paper, we propose an analysis mechanism-based structured analysis discriminative dictionary learning analysis discriminative dictionary learning, framework. The ADDL seamlessly integrates ADDL, analysis representation, and analysis classifier training into a unified model. The applied analysis mechanism can make sure that the learned dictionaries, representations, and linear classifiers over different classes are independent and discriminating as much as possible. The dictionary is obtained by minimizing a reconstruction error and an analytical incoherence promoting term that encourages the subdictionaries associated with different classes to be independent. To obtain the representation coefficients, ADDL imposes a sparse -norm constraint on the coding coefficients instead of using or norm, since the - or -norm constraint applied in most existing DL criteria makes the training phase time consuming. The code-extraction projection that bridges data with the sparse codes by extracting special features from the given samples is calculated via minimizing a sparse code approximation term. Then we compute a linear classifier based on the approximated sparse codes by an analysis mechanism to simultaneously consider the classification and representation powers. Thus, the classification approach of our model is very efficient, because it can avoid the extra time-consuming sparse reconstruction process with trained dictionary for each new test data as most existing DL algorithms. Simulations on real image databases demonstrate that our ADDL model can obtain superior performance over other state of the arts.

We propose a robust inductive semi-supervised label prediction model over the embedded representation, termed adaptive embedded label propagation with weight learning (AELP-WL), for classification. AELP-WL offers several properties. First, our method seamlessly integrates the robust adaptive embedded label propagation with adaptive weight learning into a unified framework. By minimizing the reconstruction errors over embedded features and embedded soft labels jointly, our AELP-WL can explicitly ensure the learned weights to be joint optimal for representation and classification, which differs from most existing LP models that perform weight learning separately by an independent step before label prediction. Second, existing models usually precalculate the weights over the original samples that may contain unfavorable features and noise decreasing performance. To this end, our model adds a constraint that decomposes original data into a sparse component encoding embedded noise-removed sparse representations of samples and a sparse error part fitting noise, and then performs the adaptive weight learning over the embedded sparse representations. Third, our AELP-WL computes the projected soft labels by trading-off the manifold smoothness and label fitness errors over the adaptive weights and the embedded representations for enhancing the label estimation power. By including a regressive label approximation error for simultaneous minimization to correlate sample features with the embedded soft labels, the out-of-sample issue is naturally solved. By minimizing the reconstruction errors over features and embedded soft labels, classification error and label approximation error jointly, state-of-the-art results are delivered.

Named Data Networking (NDN), a revolution of the IP architecture, provides an information-centric paradigm to transmit video traffic which occupies most of the Internet traffic. However, the data caching schemes exploited by the current NDN can easily result in the unfair distribution of cache space, which is not suitable for performance isolation when there are multiple network users and the cache space is of high utilization for video applications. In this paper, we propose a max-min utility fairness scheme for the built-in cache allocation of video applications. We propose a computing method which calculates the benefit of all video data cached in the NDN node. Based on the benefit of individual user and the relevance of video data, NDN node evicts the video data which belongs to users with the most benefits. Extensive simulation experiments have been performed to evaluate our scheme in terms of user fairness, average delay and hit ratio. The results show that our scheme can guarantee the fairness for different video Users, meanwhile, reducing the packet round trip times by improving the cache hit ratio.

During the past decade, graph-based semi-supervised learning has become one of the most important research areas in machine learning and artificial intelligence community. In this paper, we propose a Capped l2,1-Norm Regularized Dictionary Learning to construct the graph for semi-supervised learning (SSL). The new sparse coding is robust to the outliers and insensitive to noise. Then, the developed graph construction can better characterize the geometrical structure of data mainfold and be utilized for efficient graph based SSL. the proposed SSL methods can also be easily extended to deal with out-of-sample data. Simulation results show that the proposed method can achieve better performance compared with other state-of-the-art graph based SSL methods.

This paper investigates the salient machine faults representation-based classification issue by dictionary learning. A novel structured latent label consistent dictionary learning (LLC-DL) model is proposed for joint discriminative salient representation and classification. Our LLC-DL deals with the tasks by solving one objective function that aims to minimize the structured reconstruction error, structured discriminative sparse-code error and classification error simultaneously. Also, LLC-DL decomposes given signals into a sparse reconstruction part over structured latent weighted discriminative dictionary, a salient feature extraction part and an error part fitting noise. Specifically, the dictionary is learnt atom by atom, where each dictionary atom is learnt with a latent vector that reduces the disturbance between interclass atoms. The structured coding coefficients are calculated via minimizing the reconstruction error and discriminative sparse code error simultaneously. The salient representations are learnt by embedding signals onto a projection and a robust linear classifier is then trained over the learned salient features directly so that features can be ensured to be optimal for classification, where robust l2,1-norm imposed on the classifier can make the prediction results more accurate. By including a salient feature extraction term, the classification approach of LLC-DL is very efficient, since there is no need to involve an extra time-consuming sparse reconstruction process with the well-trained dictionary for each test signal. Extensive simulations versify the effectiveness of our algorithm.

We propose two nuclear- and L2,1-norm regularized 2D neighborhood preserving projection (2DNPP) methods for extracting representative 2D image features. 2DNPP extracts neighborhood preserving features by minimizing a Frobenius norm-based reconstruction error that is very sensitive noise and outliers in given data. To make the distance metric more reliable and robust, and encode the neighborhood reconstruction error more accurately, we minimize the nuclear- and L2,1-norm-based reconstruction error, respectively and measure it over each image. Technically, we propose two enhanced variants of 2DNPP, nuclear-norm-based 2DNPP and sparse reconstruction-based 2DNPP. Besides, to optimize the projection for more promising feature extraction, we also add the nuclear- and sparse L2,1-norm constraints on it accordingly, where L2,1-norm ensures the projection to be sparse in rows so that discriminative features are learnt in the latent subspace and the nuclear-norm ensures the low-rank property of features by projecting data into their respective subspaces. By fully considering the neighborhood preserving power, using more reliable and robust distance metric, and imposing the low-rank or sparse constraints on projections at the same time, our methods can outperform related state-of-the-arts in a variety of simulation settings.

We propose a new transductive label propagation method, termed Adaptive Neighborhood Propagation (Adaptive-NP) by joint L2,1-norm regularized sparse coding, for semi-supervised classification. To make the predicted soft labels more accurate for predicting the labels of samples and to avoid the tricky process of choosing the optimal neighborhood size or kernel width for graph construction, Adaptive-NP seamlessly integrates sparse coding and neighborhood propagation into a unified framework. That is, the sparse reconstruction error and classification error are combined for joint minimization, which clearly differs from traditional methods that explicitly separate graph construction and label propagation into independent steps, which may result in inaccurate predictions. Note that our Adaptive-NP alternately optimize the sparse codes and soft labels matrices, where the sparse codes are used as adaptive weights for neighborhood propagation at each iteration, so the tricky process of determining neighborhood size or kernel width is avoided. Besides, for enhancing sparse coding, we use the L2,1-norm constraint on the sparse coding coefficients and the reconstruction error at the same time for delivering more accurate and robust representations. Extensive simulations show that our model can deliver state-of-the-art performances on several public datasets for classification.

We propose a nuclear-norm regularized two-dimensional neighborhood preserving projection (2DNPP) for extracting representative 2D image features. Note that 2DNPP extracts neighborhood preserving features through minimizing the reconstruction error, but the Frobenius norm based metric is sensitive to noise and outliers. To make the distance metric more reliable and model the neighborhood reconstruction error more accurately, we impose the nuclear-norm on the neighborhood reconstruction error and measure it over each image. Technically, we propose a new variant of 2DNPP termed nuclear-norm based 2DNPP (N-2DNPP). Besides, to make delivered projection promising for feature extraction, we also include the nuclear-norm constraint on projection accordingly, where the low-rank projection can embed data into their respective subspaces. Our method can outperform related state-of-the-arts in a variety of simulation settings.

L1-norm maximization based Discriminant Locality Preserving Projection (DLPP-L1) is shown to be effective and robust to the outliers in given data, but DLPP-L1 is based on the vector space, so it has to convert those 2D matrices into high-dimensional 1D vectorized representations when handing images. But such transformation usually destroys the topology structures of images pixels, which can decrease performance. We therefore propose to extend DLPP-L1 to the 2D matrix space. A two-dimensional DLPP-L1, termed 2D-DLPP-L1, is technically proposed for image feature extraction. Compared with DLPP-L1 for representation, our proposed 2D-DLPP-L1 can effectively preserve the topology structures among image pixels in addition to inheriting the robustness property against noise and outliers. Extensive simulations on real-world image datasets show that our 2D-DLPP-L1 can deliver enhanced performance over other state-of-the-arts for recognition.

We propose a Joint nuclear norm based nonlinear Manifold Learning through linear embedding with Classification, called JMLC. By including a feature approximation error into the existing nonlinear manifold learning framework to correlate manifold features with embedded features by a linear projection, the learnt projection can handle the outside points efficiently by embedding. Besides, to encode the neighborhood reconstruction error in manifold learning part, we apply a more reliable nuclear norm based distance metric, since nuclear norm is proved to be more reliable than both L1-norm and Frobenius norm. To make learnt nonlinear features be optimal for classification so that the accuracy can be enhanced, we also minimize a robust L2,1-norm based regressive classification error over the embedded manifold features, which ensures the classification process to be robust to noise and outliers in data. Based on performing joint manifold learning and classification alternately, our JMLC can obtain a low-dimensional embedding, a linear projection and a multi-class classifier simultaneously. Extensive results demonstrate the validity of our proposed algorithm for feature extraction and robust classification, compared with other related models.

In most sparse coding based subspace clustering problems, using the non-convex lp-norm minimization (01-norm minimization. In this paper, we propose a sparse subspace clustering via joint lp-norm and l2,p-norm minimization, where the lp-norm imposed on sparse representations can achieve more sparsity for clustering while l2,p-norm imposed on reconstructed error can handle outlier pursuit. We also propose an iterative solution to solve the proposed problem based on Iterative Shrinkage/Thresholding (IST) method. In addition, to the best knowledge, utilizing IST for solving l2,p-norm minimization problem can be the first work in our paper and there is no such work before. Finally, to demonstrate the improved performance of the proposed method, comparative study was performed on benchmark problems of image clustering. Thoroughly experimental studies on real world datasets show that the method can significantly outperform other state-of-the-art methods.

We propose a new discriminative dictionary learning framework termed Latent Label Consistent K-SVD (LLC-KSVD) for representing and classifying machine faults. Our LLC-KSVD handles the task by minimizing the reconstruction, discriminative sparse-code and classification errors at the same time. To enhance the representation and classification powers, LLC-KSVD aim to decompose given data into a sparse reconstruction part, a salient feature part and an error part. The salient features are learnt by embedding data onto a projection and a classifier is then trained over extracted salient features so that features are ensured to be optimal for classification. Thus, the classification approach of our LLC-KSVD is very efficient, since there is no need to involve a time-consuming sparse reconstruction process with well-trained dictionary for each test signal as existing models. Besides, to make the classifier be robust to noise and outliers, we regularize the l2,¿-norm on classifier so that the predictions are more accurate. Simulations on several machine fault datasets demonstrate the state-of-the-art performance by our LLC-KSVD.

Wireless Sensor Networks (WSNs), the major technique in the sensing layer of Internet of Things. Because the WSN is extensively regarded as a data-centric network, more and more researchers start to exploit information-centric networking (ICN), a major branch of future network architecture, to design WSNs. In this paper, we propose a collaborative caching strategy for information-centric wireless sensor network (ICN-WSN). The proposed strategy consists of three parts: the node betweenness based cache size adjustment, the data replacement frequency based cache decision, and the content value based cache replacement algorithm. Based on our developed prototype of ICN-WSN, we compared our caching strategy with traditional strategies. Extensive experiments have been performed in terms of energy consumption, hit ratio, average delay and average hops. The results show that our strategy can improve the hit ratio of content caching, and in the meanwhile reduce the average delay, average hops, and energy consumption.

The feature extraction algorithm plays an important role in face recognition. However, the extracted features also have overlapping discriminant information. A property of the statistical uncorrelated criterion is that it eliminates the redundancy among the extracted discriminant features, while many algorithms generally ignore this property. In this paper, we introduce a novel feature extraction method called local uncorrelated local discriminant embedding (LULDE). The proposed approach can be seen as an extension of a local discriminant embedding (LDE) framework in three ways. First, a new local statistical uncorrelated criterion is proposed, which effectively captures the local information of interclass and intraclass. Second, we reconstruct the affinity matrices of an intrinsic graph and a penalty graph, which are mentioned in LDE to enhance the discriminant property. Finally, it overcomes the small-sample-size problem without using principal component analysis to preprocess the original data, which avoids losing some discriminant information. Experimental results on Yale, ORL, Extended Yale B, and FERET databases demonstrate that LULDE outperforms LDE and other representative uncorrelated feature extraction methods.

In this paper, we propose a semisupervised label consistent dictionary learning (SSDL) framework for machine fault classification. SSDL is a semisupervised extension of recent fully supervised label consistent dictionary learning approach, since the number of labeled machine data is usually limited in practice. To enable the supervised dictionary learning model to use both labeled and commonly readily available unlabeled data for enhancing performance, we propose to incorporate the merits of label prediction and present a joint label consistent dictionary learning and adaptive label prediction technique. In this setting, we first employ the existing label prediction model to estimate the labels of unlabeled training signals in a transductive fashion for enriching supervised prior. Then, we use predicted labeled data for label consistent dictionary learning. After that, we apply the discriminant sparse codes as the adaptive reconstruction weights for label prediction to update the estimated labels of unlabeled training data and the discriminative sparse codes matrix for label consistent dictionary learning so that classification performance can be enhanced. Thus, an informative dictionary, a sparse-code matrix, and an optimal multiclass classifier can be alternately obtained from one objective function. Besides, the tricky process of choosing optimal kernel width and neighborhood size can also be effectively voided in our scheme due to the adaptive weights. Extensive simulations on several machine fault datasets show that our SSDL method can deliver enhanced performance over other state-of-the-arts for machine fault classification.

Recovering low-rank and sparse subspaces jointly for enhanced robust representation and classification is discussed. Technically, we first propose a transductive low-rank and sparse principal feature coding (LSPFC) formulation that decomposes given data into a component part that encodes low-rank sparse principal features and a noise-fitting error part. To well handle the outside data, we then present an inductive LSPFC (I-LSPFC). I-LSPFC incorporates embedded low-rank and sparse principal features by a projection into one problem for direct minimization, so that the projection can effectively map both inside and outside data into the underlying subspaces to learn more powerful and informative features for representation. To ensure that the learned features by I-LSPFC are optimal for classification, we further combine the classification error with the feature coding error to form a unified model, discriminative LSPFC (D-LSPFC), to boost performance. The model of D-LSPFC seamlessly integrates feature coding and discriminative classification, so the representation and classification powers can be enhanced. The proposed approaches are more general, and several recent existing low-rank or sparse coding algorithms can be embedded into our problems as special cases. Visual and numerical results demonstrate the effectiveness of our methods for representation and classification.

Semisupervised dimension reduction via virtual label regression first derives the virtual labels of unlabeled data by employing a newly designed label propagation (LP) approach (called Special random walk (SRW)) and then encodes them in a weighted linear regression model. Nie et al. (2011) highlighted two important characteristics of SRW nonexistent in the previous LP approaches: outlier detection and probability value output, which guarantee the elegant encoding of the resultant virtual labels in the weighted label regression. However, in this brief, we show that the relationship between the SRW and the previous work on LP is very close. Naturally, a problem deserving investigation is whether traditional LP approaches are indeed unable to share the above two characteristics of SRW. We aim to address this problem.

Named Data Networking (NDN), a revolution of the IP architecture, provides an information-centric routing for designing the unified Internet of Things (IoT) network protocol. However, the data forwarding scheme exploited by the current NDN generates lots of communication overhead and invalid caching hits, which is not suitable for IoT. This is because there exist large amounts of weak network devices with constrained resources in IoT compared to the Internet. In this paper, we first build a caching time model to characterize the caching process for individual packet and estimate the packet's caching time on the next hop node. Based on this model, we propose a cache-aware named-data forwarding scheme for the typical weak devices in IoT-wireless sensor nodes. Extensive simulation experiments have been performed to evaluate our scheme in terms of forwarding number, average delay and valid hit ratio. The results show that our scheme can reduce the packet number during multicast forwarding of router nodes, meanwhile guarantee valid hit ratio of content caching.

In this paper, we propose a L1-Norm driven Semi-Supervised Local Discriminant Projection (S2LDP-L1) for robust dimensionality reduction and image representation. For feature learning, our S2LDP-L1 approach aims at compacting local within-class divergence and separating local betweenclass divergence at the same time in addition to possessing the locality preserving power over all training data. To enable the presented S2LDP-L1 method to be robust against noise in data for feature reduction and representation, the L1-norm that is proven to be robust to noise and outliers is regularized on the constructed scatter matrices for measuring pairwise similarities/ dissimilarities between samples. Thus, the presentation power can be effectively enhanced to improve the subsequent classification task. The derived ratio based model is finally solved by an iterative approach to deliver a discriminating and neighborhood preserving orthogonal projection for extracting features from both training and test samples by embedding data onto it. For classification, an existing label propagation model is used to identify the categories of test data. Extensive results on handwriting digit datasets verified the validity of our S2LDP-L1, compared with other state-of-the-arts.

1-norm support vector machine (SVM) has attracted substantial attentions for its good sparsity. However, the computational complexity of training 1-norm SVM is about the cube of the sample number, which is high. This paper replaces the hinge loss or the ε-insensitive loss by the squared loss in the 1-norm SVM, and applies orthogonal matching pursuit (OMP) to approximate the solution of the 1-norm SVM with the squared loss. Experimental results on toy and real-world datasets show that OMP can faster train 1-norm SVM and achieve similar learning performance compared with some methods available.

In this paper, we mainly present a Semi-Supervised Label Consistent KSVD (S2KSVD) algorithm for representing and classifying machine faults. The formulation of our S2KSVD is an improvement to the recent label consistent K-SVD (LC-KSVD), because LC-KSVD is a fully supervised approach, and needs to use supervised class information of all training data to compute a reconstructive & discriminative dictionary. But labeled signals are often expensive to obtain, while in contrast unlabeled signals can be easily captured with low expense from the real world. Thus, the application of LC-KSVD may be constrained in reality. To address this problem, we present S2KSVD through involving a computationally efficient label propagation (LP) process as a preprocessing step. The core idea is to employ the LP process to estimate the labels of unlabeled signals so that supervised prior knowledge that can significantly enhance classification can be increased. Simulation results on several machine fault datasets demonstrate that our algorithm delivers promising performance for machine fault classification.

This paper proposes an enhanced semi-supervised classification approach termed Nonnegative Sparse Neighborhood Propagation (SparseNP) that is an improvement to the existing neighborhood propagation due to the fact that the outputted soft labels of points cannot be ensured to be sufficiently sparse, discriminative, robust to noise and be probabilistic values. Note that the sparse property and strong discriminating ability of predicted labels is important, since ideally the soft label of each sample should have only one or few positive elements (that is, less unfavorable mixed signs are included) deciding its class assignment. To reduce the negative effects of unfavorable mixed signs on the learning performance, we regularize the l2,1-norm on the soft labels during optimization for enhancing the prediction results. The non-negativity and sum-to-one constraints are also included to ensure the outputted labels are probabilistic values. The proposed framework is solved in an alternative manner for delivering a more reliable solution so that the accuracy can be improved. Simulations show that satisfactory results can be obtained by the proposed SparseNP compared with other related approaches.

Traditional label propagation (LP) is shown to be effective for transductive classification. To enable the standard LP to handle outside images, two inductive methods by label reconstruction or by direct embedding have been presented, of which the latter scheme is relatively more efficient, especially for testing. But almost all inductive LP models use 1D vectors of images as inputs, which may destroy the topology structure of image pixels and usually suffer from high complexity due to the high dimension of 1D vectors in reality. To preserve the topology among pixels and address the scalability issue for the embedding based scheme, we propose a simple yet efficient Bilinear Embedding Label Propagation (BELP) by including a bilinear regularization term in terms of tensor representation to correlate the image labels with their bilinear features. BELP performs label prediction over the 2D matrices rather than 1D vectors, since images are essentially matrices. Finally, labels of new images can be easily obtained by embedding them onto a spanned bilinear subspace solved from a joint framework. Simulations verified the efficiency of our approach.

In this article, we consider the problem of simultaneous low-rank recovery and sparse projection. More specifically, a new Robust Principal Component Analysis (RPCA)-based framework called Sparse Projection and Low-Rank Recovery (SPLRR) is proposed for handwriting representation and salient stroke feature extraction. In addition to achieving a low-rank component encoding principal features and identify errors or missing values from a given data matrix as RPCA, SPLRR also learns a similarity-preserving sparse projection for extracting salient stroke features and embedding new inputs for classification. These properties make SPLRR applicable for handwriting recognition and stroke correction and enable online computation. A cosine-similarity-style regularization term is incorporated into the SPLRR formulation for encoding the similarities of local handwriting features. The sparse projection and low-rank recovery are calculated from a convex minimization problem that can be efficiently solved in polynomial time. Besides, the supervised extension of SPLRR is also elaborated. The effectiveness of our SPLRR is examined by extensive handwritten digital repairing, stroke correction, and recognition based on benchmark problems. Compared with other related techniques, SPLRR delivers strong generalization capability and state-of-the-art performance for handwriting representation and recognition.

A new graph based constrained semi-supervised learning (G-CSSL) framework is proposed. Pairwise constraints (PC) are used to specify the types (intra- or inter-class) of points with labels. Since the number of labeled data is typically small in SSL setting, the core idea of this framework is to create and enrich the PC sets using the propagated soft labels from both labeled and unlabeled data by special label propagation (SLP), and hence obtaining more supervised information for delivering enhanced performance. We also propose a Two-stage Sparse Coding, termed TSC, for achieving adaptive neighborhood for SLP. The first stage aims at correcting the possible corruptions in data and training an informative dictionary, and the second stage focuses on sparse coding. To deliver enhanced inter-class separation and intra-class compactness, we also present a mixed soft-similarity measure to evaluate the similarity/dissimilarity of constrained pairs using the sparse codes and outputted probabilistic values by SLP. Simulations on the synthetic and real datasets demonstrated the validity of our algorithms for data representation and image recognition, compared with other related state-of-the-art graph based semi-supervised techniques.

An enhanced label propagation technique termed transformed neighborhood propagation (TNP) is proposed for semi-supervised learning. In the TNP setting, the processes of constructing weighted similarity graph and propagating label information of the labeled data to unlabeled points are conducted in the transformed feature space. TNP is mainly motivated by a fact that the optimal feature representation Y with possible unfavorable features and noises in the original data X removed by feature learning are more appropriate and accurate for measuring pair wise similarities of samples. To achieve the representation Y, The recent marginal semi-supervised sub-manifold projections is applied, so enhanced inter-class separation and enhanced intra-class compactness are delivered at the same time. The similarity graph is finally constructed based on Y. We also propose to calculate semi-supervised reconstruction weights for the weight assignment. As a result, the label estimation power can be enhanced by benefiting from the refined weighted similarity graph over Y instead of X, through propagating the labels of points in the transformed space for prediction. Visualization and image classification verified the effectiveness of our TNP, compared with other related label propagation algorithms.

For low-rank recovery and error correction, Low-Rank Representation (LRR) row-reconstructs given data matrix X by seeking a low-rank representation, while Inductive Robust Principal Component Analysis (IRPCA) aims to calculate a low-rank projection to column-reconstruct X. But either column or row information of Xis lost by LRR and IRPCA. In addition, the matrix X itself is chosen as the dictionary by LRR, but (grossly) corrupted entries may greatly depress its performance. To solve these issues, we propose a simultaneous low-rank representation and dictionary learning framework termed Tensor LRR (TLRR) for robust bilinear recovery. TLRR reconstructs given matrix X along both row and column directions by computing a pair of low-rank matrices alternately from a nuclear norm minimization problem for constructing a low-rank tensor subspace. As a result, TLRR in the optimizations can be regarded as enhanced IRPCA with noises removed by low-rank representation, and can also be considered as enhanced LRR with a clean informative dictionary using a low-rank projection. The comparison with other criteria shows that TLRR exhibits certain advantages, for instance strong generalization power and robustness enhancement to the missing values. Simulations verified the validity of TLRR for recovery.

This paper discusses a new setting of graph based semi-supervised learning (SSL) guided using pairwise constraints (PCs). Technically, we propose a novel Graph based Constrained Semi-Supervised Learning (G-CSSL) framework. In this setting, PCs are used to specify the types (intra- or inter-class) of points with labels. Because the number of labeled data is typically small in SSL setting, the core idea of this framework is to create and enrich the PCs sets using the propagated soft labels from both labeled and unlabeled data via special label propagation (SLP), and hence obtaining more supervised information for delivering enhanced learning performance. To obtain the predicted labels of unlabeled data, we calculate the sparse codes of all data vectors jointly to assign weights for SLP. To deliver enhanced inter-class separation and intra-class compactness, we also present a mixed soft-similarity measure to evaluate the similarity/dissimilarity of constrained sample pairs by using the sparse codes and outputted probabilistic values by SLP. Extensive simulations demonstrated the effectiveness of our G-CSSL for image representation and recognition, compared with other related SSL techniques.

This paper proposes a Robust Principal Component Analysis (RPCA) based framework called Sparse Projection and Low-Rank Recovery (SPLRR) for representing and recognizing handwritings. SPLRR calculates a similarity preserving sparse projection for salient feature extraction and processing new data for classification in addition to delivering a low-rank principal component and identifying errors or missing pixel values from a given data matrix. As a result, SPLRR will be applicable for handwritten recovery, recognition and the applications requiring online computation. To encode the similarity between features in the learning process, the Cosine similarity based regularization is incorporated to the SPLRR formulation. The sparse projection and the lowest-rank components are calculated from a scalable convex minimization problem that can be efficiently solved in polynomial time. The effectiveness of the proposed SPLRR is examined by handwritten digital repairing, stroke correction and recognition on two real problems. Results show that SPLRR can deliver state-of-the-art results in handwriting representation.

This paper discusses the image representation and decomposition problem using enhanced low-rank representation. Technically, we propose a Regularized Low-Rank Representation framework referred to as rLRR that is motivated by the fact that Latent Low-Rank Representation (LatLRR) delivers robust and promising results for image representation and feature extraction through recovering the hidden effects, but the locality among both similar principal and salient features to be encoded are not preserved in the original LatLRR formulation. To address this problem for obtaining enhanced performance, rLRR is proposed through incorporating an appropriate Laplacian regularization term that allows us to keep the local geometry of close features. Similar to LatLRR, rLRR decomposes a given data matrix from two directions by calculating a pair of low-rank matrices. But the similarities among principal features and salient features can be clearly preserved by rLRR. Thus the correlated features can be well grouped and the robustness of representations can also be effectively improved. The effectiveness of rLRR is examined by representation and recognition of real images. Results verified the validity of our presented rLRR technique.

This paper introduces an effective mechanism to extract informative principal features from the color images and proposes a color principal feature extraction technique referred to as ColorPCA. ColorPCA performs in the color image space, extracting the principal features directly from the color images. As a result, the color and local topological information of pixels at each level of the color images can be effectively preserved. In extracting the most representative features, a color image scatter matrix is constructed and its eigenvectors are employed for color principal feature extraction. ColorPCA has only one parameter (i.e., the reduced dimension) to estimate and the projection axes can be effectively obtained using eigen-decomposition. Extensive color image reconstruction and recognition over the benchmark problems verified the effectiveness of the presented ColorPCA. Results show that ColorPCA can effectively reconstruct the color images. Image recognition also demonstrates that ColorPCA can deliver promising results compared with other state-of-the-art 1D and 2D principal feature extraction algorithms.

Semi-supervised dimensionality reduction is an important research topic in many pattern recognition and machine learning applications. Among all the methods for semi-supervised dimensionality reduction, SDA and LapRLS are two popular ones. Though the two methods are actually the extensions of different supervised methods, we show in this paper that they can be unified into a regularized least square framework. However, the regularization term added to the framework focuses on smoothing only, it cannot fully utilize the underlying discriminative information which is vital for classification. In this paper, we propose a new effective semi-supervised dimensionality reduction method, called LLGDI, to solve the above problem. The proposed LLGDI method introduces a discriminative manifold regularization term by using the local discriminative information instead of only relying on neighborhood information. In this way, both the local geometrical and discriminative information of dataset can be preserved by the proposed LLGDI method. Theoretical analysis and extensive simulations show the effectiveness of our algorithm. The results in simulations demonstrate that our proposed algorithm can achieve great superiority compared with other existing methods.

Dealing with high-dimensional data has always been a major problem with the research of pattern recognition and machine learning. And Linear Discriminant Analysis (LDA) is one of the most popular methods for dimensionality reduction. But it only uses labeled samples while neglect the unlabeled samples, which are abundant and can be easily obtained in the real world. In this paper, we propose a new dimensionality reduction method by using the unlabeled samples to enhance the performance of LDA. The new method first propagates the label information from labeled set to unlabeled set via a label propagation process, where the predicted labels of unlabeled samples, called soft labels, can be obtained. It then incorporates the soft labels into the construction of scatter matrixes to find a transformed matrix for dimensionality reduction. In this way, the proposed method can preserve more discriminative information, which is good when solving the classification problem. Extensive simulations are carried based several datasets and the results show the effectiveness of the proposed method.

The problem of learning from both labeled and unlabeled data is considered. In this paper, we present a novel semisupervised multimodal dimensionality reduction (SSMDR) algorithm for feature reduction and extraction. SSMDR can preserve the local and multimodal structures of labeled and unlabeled samples. As a result, data pairs in the close vicinity of the original space are projected in the nearby of the embedding space. Due to overfitting, supervised dimensionality reduction methods tend to perform inefficiently when only few labeled samples are available. In such cases, unlabeled samples play a significant role in boosting the learning performance. The proposed discriminant technique has an analytical form of the embedding transformations that can be effectively obtained by applying the eigen decomposition, or finding two close optimal sets of transforming basis vectors. By employing the standard kernel trick, SSMDR can be extended to the nonlinear dimensionality reduction scenarios. We verify the feasibility and effectiveness of SSMDR through conducting extensive simulations including data visualization and classification on the synthetic and real‐world datasets. Our obtained results reveal that SSMDR offers significant advantages over some widely used techniques. Compared with other methods, the proposed SSMDR exhibits superior performance on multimodal cases.

Isomap is a well-known nonlinear dimensionality reduction (DR) method, aiming at preserving geodesic distances of all similarity pairs for delivering highly nonlinear manifolds. Isomap is efficient in visualizing synthetic data sets, but it usually delivers unsatisfactory results in benchmark cases. This paper incorporates the pairwise constraints into Isomap and proposes a marginal Isomap (M-Isomap) for manifold learning. The pairwise Cannot-Link and Must-Link constraints are used to specify the types of neighborhoods. M-Isomap computes the shortest path distances over constrained neighborhood graphs and guides the nonlinear DR through separating the interclass neighbors. As a result, large margins between both interand intraclass clusters are delivered and enhanced compactness of intracluster points is achieved at the same time. The validity of M-Isomap is examined by extensive simulations over synthetic, University of California, Irvine, and benchmark real Olivetti Research Library, YALE, and CMU Pose, Illumination, and Expression databases. The data visualization and clustering power of M-Isomap are compared with those of six related DR methods. The visualization results show that M-Isomap is able to deliver more separate clusters. Clustering evaluations also demonstrate that M-Isomap delivers comparable or even better results than some state-of-the-art DR algorithms.

Two novel unsupervised dimensionality reduction techniques, termed sparse distance preserving embedding (SDPE) and sparse proximity preserving embedding (SPPE), are proposed for feature extraction and classification. SDPE and SPPE perform in the clean data space recovered by sparse representation and enhanced Euclidean distances over noise removed data are employed to measure pairwise similarities of points. In extracting informative features, SDPE and SPPE aim at preserving pairwise similarities between data points in addition to preserving the sparse characteristics. This paper calculates the sparsest representation of all vectors jointly by a convex optimization. The sparsest codes enable certain local information of data to be preserved, and can endow SDPE and SPPE a natural discriminating power, adaptive neighborhood and robust characteristic against noise and errors in delivering low-dimensional embeddings. We also mathematically show SDPE and SPPE can be effectively extended for discriminant learning in a supervised manner. The validity of SDPE and SPPE is examined by extensive simulations. Comparison with other related state-of-the-art unsupervised algorithms show that promising results are delivered by our techniques.

Visualizing similarity data of different objects by exhibiting more separate organizations with local and multimodal characteristics preserved is important in multivariate data analysis. Laplacian Eigenmaps (LAE) and Locally Linear Embedding (LLE) aim at preserving the embeddings of all similarity pairs in the close vicinity of the reduced output space, but they are unable to identify and separate interclass neighbors. This paper considers the semi-supervised manifold learning problems. We apply the pairwise Cannot-Link and Must-Link constraints induced by the neighborhood graph to specify the types of neighboring pairs. More flexible regulation on supervised information is provided. Two novel multimodal nonlinear techniques, which we call trace ratio (TR) criterion-based semi-supervised LAE (S2LAE) and LLE (S2LLE), are then proposed for marginal manifold visualization. We also present the kernelized S2LAE and S2LLE. We verify the feasibility of S2LAE and S2LLE through extensive simulations over benchmark real-world MIT CBCL, CMU PIE, MNIST, and USPS data sets. Manifold visualizations show that S2LAE and S2LLE are able to deliver large margins between different clusters or classes with multimodal distributions preserved. Clustering evaluations show they can achieve comparable to or even better results than some widely used methods.

This paper incorporates the group sparse representation into the well-known canonical correlation analysis (CCA) framework and proposes a novel discriminant feature extraction technique named group sparse canonical correlation analysis (GSCCA). GSCCA uses two sets of variables and aims at preserving the group sparse (GS) characteristics of data within each set in addition to maximize the global interset covariance. With GS weights computed prior to feature extraction, the locality, sparsity and discriminant information of data can be adaptively determined. The GS weights are obtained from an NP-hard group-sparsity promoting problem that considers all highly correlated data within a group. By defining one of the two variable sets as the class label matrix, GSCCA is effectively extended to multiclass scenarios. Then GSCCA is theoretically formulated as a least-squares problem as CCA does. Comparative analysis between this work and the related studies demonstrate that our algorithm is more general exhibiting attractive properties. The projection matrix of GSCCA is analytically solved by applying eigen-decomposition and trace ratio (TR) optimization. Extensive benchmark simulations are conducted to examine GSCCA. Results show that our approach delivers promising results, compared with other related algorithms.

Semi-supervised dimensionality reduction is an important research topic in many pattern recognition and machine learning applications. Among all the methods for semi-supervised dimensionality reduction, SDA and Lap-LDA are two popular ones. Both SDA and Lap-LDA can perform dimensionality reduction by preserving the discriminative structure embedding in the labeled samples as well as the manifold structure embedded both in labeled and unlabeled samples. But they apply different schemes for semi-supervised dimensionality reduction. SDA has added the manifold term to the objective function of LDA while Lap-LDA has added such term to the objective function of Least Square with certain class indicator. In this paper, we further analyze the schemes of two methods and build the equivalence between them by giving a certain condition. We then show their difference when the certain condition cannot be satisfied. Extensive simulations have been conducted based several datasets. Both theoretical analysis and simulation results confirm the analysis. Finally, motivated by the equivalence and differences between two methods, we then propose an improved approach for semi-supervised dimensionality reduction. The proposed approach is actually a two-stage approach and can obtain the optimal solution equivalent to Lap-LDA (in this first stage) and SDA (in the second stage) with less computational cost.

This paper discusses the semi-supervised marginal projection problems learning from partial constrained data. Two effective multimodal dimensionality reduction (DR) algorithms, which we call semi-supervised marginal projections (SSMP) and orthogonal SSMP (OSSMP), are proposed. By specifying the types of similarity pairs with the pairwise constraints (PC), our techniques can preserve the global structures of all points as well as local geometrical and discriminant structures embedded in the PC. SSMP in singular case is also discussed. Because in all the PC guided methods, extracting the informative constraints is difficult and random constraints greatly affect the learning performance of techniques, this work also presents an effective and efficient methodology of optimally selecting the informative constraints for learning. The analytic form of the marginal projections can be effectively obtained by eigen-decomposition. The connections between this present work and the related semi-supervised algorithms are also detailed. The effectiveness of our proposed informative constraint selection method and algorithms are evaluated by benchmark problems. Results show our methods are capable of delivering competitive results with some widely used state-of-the-art semi-supervised algorithms.

Linear Discriminant Analysis (LDA) is a well-known dimensionality reduction algorithm for pattern recognition and machine learning. And Trace Ratio LDA (TR-LDA) and Null-space LDA (NLDA) are two popular variants of LDA. Both NLDA and TR-LDA can result in orthogonal transformations. However, they applied different schemes in deriving the optimal transformation. NLDA computes an orthogonal transformation in the null space of the within-class scatter matrix, while TRLDA computes an orthogonal transformation by an iterative procedure. In this paper, by using the trace difference problem as a bridge, we show that the above two algorithms can be equivalent when confronts with singularity problem. In addition, extensive simulations were conducted based on several datasets. Both theoretical analysis and simulation results confirm the equivalent relationship.

Dimensionality reduction has been a fundamental tool when dealing with high-dimensional dataset. And trace ration optimization has been widely used in dimensionality reduction because Trace ratio can directly reflect the similarity (Euclidean distance) of data points. Conventionally, there is no close-form solution to the original trace ratio problem. Prior works have indicated that trace ratio problem can be solved by an iterative way. In this paper, we propose an efficient algorithm to find the optimal solutions. The proposed algorithm can be easily extended to its corresponding kernel version for handling the nonlinear problems. Finally, we evaluate our proposed algorithm based on extensive simulations of real world datasets. The results show our proposed method is able to deliver marked improvements over other supervised and unsupervised algorithms.

This letter presents a Tensor Locally Linear Discriminative Analysis (TLLDA) method for image presentation. TLLDA is originated from the Local Fisher Discriminant Analysis (LFDA), but TLLDA offers some advantages over LFDA. 1) TLLDA can preserve the local discriminative information of image data as LFDA. 2) TLLDA represents images as matrices or 2-order tensors rather than vectors, so TLLDA keeps the spatial locality of pixels in the images. 3) TLLDA avoids the singularity that may be suffered by LFDA. 4) TLLDA is faster than LFDA. Simulations on two real databases verified the validity of TLLDA. Results show that TLLDA is highly competitive with some widely used techniques.