The introduction of diffusion models has brought significant advances to the field of audio-driven talking head generation. However, the extremely slow inference speed severely limits the practical implementation of diffusion-based talking head generation models. In this study, we propose READ, a real-time diffusion-transformer-based talking head generation framework. Our approach first learns a spatiotemporal highly compressed video latent space via a temporal VAE, significantly reducing the token count to accelerate generation. To achieve better audio-visual alignment within this compressed latent space, a pre-trained Speech Autoencoder (SpeechAE) is proposed to generate temporally compressed speech latent codes corresponding to the video latent space. These latent representations are then modeled by a carefully designed Audio-to-Video Diffusion Transformer (A2V-DiT) backbone for efficient talking head synthesis. Furthermore, to ensure temporal consistency and accelerated inference in extended generation, we propose a novel asynchronous noise scheduler (ANS) for both the training and inference processes of our framework. The ANS leverages asynchronous add-noise and asynchronous motion-guided generation in the latent space, ensuring consistency in generated video clips. Experimental results demonstrate that READ outperforms state-of-the-art methods by generating competitive talking head videos with significantly reduced runtime, achieving an optimal balance between quality and speed while maintaining robust metric stability in long-time generation.

3D Gaussian Splatting (3D-GS) has emerged as an efficient 3D representation and a promising foundation for semantic tasks like segmentation. However, existing 3D-GS-based segmentation methods typically rely on high-dimensional category features, which introduce substantial memory overhead. Moreover, fine-grained segmentation remains challenging due to label space congestion and the lack of stable multi-granularity control mechanisms. To address these limitations, we propose a coarse-to-fine binary encoding scheme for per-Gaussian category representation, which compresses each feature into a single integer via the binary-to-decimal mapping, drastically reducing memory usage. We further design a progressive training strategy that decomposes panoptic segmentation into a series of independent sub-tasks, reducing inter-class conflicts and thereby enhancing fine-grained segmentation capability. Additionally, we fine-tune opacity during segmentation training to address the incompatibility between photometric rendering and semantic segmentation, which often leads to foreground-background confusion. Extensive experiments on multiple benchmarks demonstrate that our method achieves state-of-the-art segmentation performance while significantly reducing memory consumption and accelerating inference.

Existing Transformer-based RGBT trackers achieve remarkable performance benefits by leveraging self-attention to extract uni-modal features and cross-attention to enhance multi-modal feature interaction and search-template correlation. Nevertheless, the independent search-template correlation calculations are prone to be affected by low-quality data, which might result in contradictory and ambiguous correlation weights. It not only limits the intra-modal feature representation, but also harms the robustness of cross-attention for multi-modal feature interaction and search-template correlation computation. To address these issues, we propose a novel approach called Cross-modulated Attention Transformer (CAFormer), which innovatively integrates inter-modality interaction into the search-template correlation computation within typical attention mechanism, for RGBT tracking. In particular, we first independently generate correlation maps for each modality and feed them into the designed correlation modulated enhancement module, which can modify inaccurate correlation weights by seeking the consensus between modalities. Such kind of design unifies self-attention and cross-attention schemes, which not only alleviates inaccurate attention weight computation in self-attention but also eliminates redundant computation introduced by extra cross-attention scheme. In addition, we design a collaborative token elimination strategy to further improve tracking inference efficiency and accuracy. Experiments on five public RGBT tracking benchmarks show the outstanding performance of the proposed CAFormer against state-of-the-art methods.

Diffusion models have revolutionized the field of talking head generation, yet still face challenges in expressiveness, controllability, and stability in long-time generation. In this research, we propose an EmotiveTalk framework to address these issues. Firstly, to realize better control over the generation of lip movement and facial expression, a Vision-guided Audio Information Decoupling (V-AID) approach is designed to generate audio-based decoupled representations aligned with lip movements and expression. Specifically, to achieve alignment between audio and facial expression representation spaces, we present a Diffusion-based Co-speech Temporal Expansion (Di-CTE) module within V-AID to generate expression-related representations under multi-source emotion condition constraints. Then we propose a well-designed Emotional Talking Head Diffusion (ETHD) backbone to efficiently generate highly expressive talking head videos, which contains an Expression Decoupling Injection (EDI) module to automatically decouple the expressions from reference portraits while integrating the target expression information, achieving more expressive generation performance. Experimental results show that EmotiveTalk can generate expressive talking head videos, ensuring the promised controllability of emotions and metric stability during long-time generation, yielding state-of-the-art performance compared to existing methods. The main page of our paper can be found in https://emotivetalk.github.io/.

Recently, visual-language learning (VLL) has shown great potential in enhancing visual-based person re-identification (ReID). Existing VLL-based ReID methods typically focus on image-text feature alignment at the whole-body level, while neglecting supervision on fine-grained part features, thus lacking constraints for local feature semantic consistency. To this end, we propose Part-Informed Visual-language Learning (π-VL) to enhance fine-grained visual features with part-informed language supervisions for ReID tasks. Specifically, π-VL introduces a human parsing-guided prompt tuning strategy and a hierarchical visual-language alignment paradigm to ensure within-part feature semantic consistency. The former combines both identity labels and human parsing maps to constitute pixel-level text prompts, and the latter fuses multi-scale visual features with a light-weight auxiliary head to perform fine-grained image-text alignment. As a plug-and-play and inference-free solution, our π-VL achieves performance comparable to or better than state-of-the-art methods on four commonly used ReID benchmarks. Notably, it reports 91.0% Rank-1 and 76.9% mAP on the challenging MSMT17 database, without bells and whistles.

In the field of historical manuscript research, scholars frequently encounter novel symbols in ancient texts, investing considerable effort in their identification and documentation. Although existing object detection methods achieve impressive performance on known categories, they struggle to recognize novel symbols without retraining. To address this limitation, we propose a Visually Guided Text Spotting (VGTS) approach that accurately spots novel characters using just one annotated support sample. The core of VGTS is a spatial alignment module consisting of a Dual Spatial Attention (DSA) block and a Geometric Matching (GM) block. The DSA block aims to identify, focus on, and learn discriminative spatial regions in the support and query images, mimicking the human visual spotting process. It first refines the support image by analyzing inter-channel relationships to identify critical areas, and then refines the query image by focusing on informative key points. The GM block, on the other hand, establishes the spatial correspondence between the two images, enabling accurate localization of the target character in the query image. To tackle the example imbalance problem in low-resource spotting tasks, we develop a novel torus loss function that enhances the discriminative power of the embedding space for distance metric learning . To further validate our approach, we introduce a new dataset featuring ancient Dongba hieroglyphics (DBH) associated with the Naxi minority of China. Extensive experiments on the DBH dataset and other public datasets, including Egyptian Hieroglyph (EGY), Historical Arabic Documents (HAD), Tripitaka Koreana in Han (TKH), and Notary Charters (NC), show that VGTS consistently surpasses state-of-the-art methods. The proposed framework exhibits great potential for application in historical manuscript text spotting, enabling scholars to efficiently identify and document novel symbols with minimal annotation effort.

The Segment Anything Model (SAM), a profound vision foundation model pretrained on a large-scale dataset, breaks the boundaries of general segmentation and sparks various downstream applications. This paper introduces Hi-SAM, a unified model leveraging SAM for hierarchical text segmentation. Hi-SAM excels in segmentation across four hierarchies, including pixel-level text, word, text-line, and paragraph, while realizing layout analysis as well. Specifically, we first turn SAM into a high-quality pixel-level text segmentation (TS) model through a parameter-efficient fine-tuning approach. We use this TS model to iteratively generate the pixel-level text labels in a semi-automatical manner, unifying labels across the four text hierarchies in the HierText dataset. Subsequently, with these complete labels, we launch the end-to-end trainable Hi-SAM based on the TS architecture with a customized hierarchical mask decoder. During inference, Hi-SAM offers both automatic mask generation (AMG) mode and promptable segmentation (PS) mode. In the AMG mode, Hi-SAM segments pixel-level text foreground masks initially, then samples foreground points for hierarchical text mask generation and achieves layout analysis in passing. As for the PS mode, Hi-SAM provides word, text-line, and paragraph masks with a single point click. Experimental results show the state-of-the-art performance of our TS model: 84.86% fgIOU on Total-Text and 88.96% fgIOU on TextSeg for pixel-level text segmentation. Moreover, compared to the previous specialist for joint hierarchical detection and layout analysis on HierText, Hi-SAM achieves significant improvements: 4.73% PQ and 5.39% F1 on the text-line level, 5.49% PQ and 7.39% F1 on the paragraph level layout analysis, requiring fewer training epochs.

Watermark is essential for protecting the intellectual property of private images. However, a wide range of watermark removal attacks, especially many AI-powered ones, can automatically predict and remove watermarks, posing serious concerns. In this paper, we present the design ofImprints, a defensive watermarking framework that fortifies watermarks against watermark removal attacks. By formulating an optimization problem that deters watermark removal attacks, we design image-independent/dependent defensive watermark models for effective batch/customized protection. We further enhance the watermark to be transferable to unseen watermark removal attacks and robust to editing distortions. Extensive experiments verify thatImprintsoutperforms existing baselines in terms of its immunity to 8 state-of-the-art watermark removal attacks and 3 commercial black-box watermark removal software. The source code is available athttps://github.com/Imprints-wm/Imprints.

Scene text recognition is inherently a vision-language task. However, previous works have predominantly focused either on extracting more robust visual features or designing better language modeling. How to effectively and jointly model vision and language to mitigate heavy reliance on a single modality remains a problem. In this paper, aiming to enhance vision-language reasoning in scene text recognition, we present a balanced, unified and synchronized vision-language reasoning network (BUSNet). Firstly, revisiting the image as a language by balanced concatenation along length dimension alleviates the issue of over-reliance on vision or language. Secondly, BUSNet learns an ensemble of unified external and internal vision-language model with shared weight by masked modality modeling (MMM). Thirdly, a novel vision-language reasoning module (VLRM) with synchronized vision-language decoding capacity is proposed. Additionally, BUSNet achieves improved performance through iterative reasoning, which utilizes the vision-language prediction as a new language input. Extensive experiments indicate that BUSNet achieves state-of-the-art performance on several mainstream benchmark datasets and more challenge datasets for both synthetic and real training data compared to recent outstanding methods. Code and dataset will be available at https://github.com/jjwei66/BUSNet.

Abstract Due to the huge cost of manual annotations, the labelled data may not be sufficient to train a dynamic facial expression (DFR) recogniser with good performance. To address this, the authors propose a multi‐modal pre‐training method with a pseudo‐label guidance mechanism to make full use of unlabelled video data for learning informative representations of facial expressions. First, the authors build a pre‐training dataset of videos with aligned vision and audio modals. Second, the vision and audio feature encoders are trained through an instance discrimination strategy and a cross‐modal alignment strategy on the pre‐training data. Third, the vision feature encoder is extended as a dynamic expression recogniser and is fine‐tuned on the labelled training data. Fourth, the fine‐tuned expression recogniser is adopted to predict pseudo‐labels for the pre‐training data, and then start a new pre‐training phase with the guidance of pseudo‐labels to alleviate the long‐tail distribution problem and the instance‐class confliction. Fifth, since the representations learnt with the guidance of pseudo‐labels are more informative, a new fine‐tuning phase is added to further boost the generalisation performance on the DFR recognition task. Experimental results on the Dynamic Facial Expression in the Wild dataset demonstrate the superiority of the proposed method.

Zero-shot cross-lingual information extraction (IE) aims at constructing an IE model for some low-resource target languages, given annotations exclusively in some rich-resource languages. Recent studies have shown language-universal features can bridge the gap between languages. However, prior work has neither explored the potential of establishing interactions between language-universal features and contextual representations nor incorporated features that can effectively model constituent span attributes and relationships between multiple spans. In this study, asyntax-augmentedhierarchicalinteractiveencoder (SHINE) is proposed to transfer cross-lingual IE knowledge. The proposed encoder is capable of interactively capturing complementary information between features and contextual information, to derive language-agnostic representations for various cross-lingual IE tasks. Concretely, a multi-level interaction network is designed to hierarchically interact the complementary information to strengthen domain adaptability. Besides, in addition to the well-studied word-level syntax features of part-of-speech and dependency relation, a new span-level syntax feature of constituency structure is introduced to model the constituent span information which is crucial for IE. Experiments across seven languages on three IE tasks and four benchmarks verify the effectiveness and generalization ability of the proposed method.

The problem of document structure reconstruction refers to converting digital or scanned documents into corresponding semantic structures. Most existing works mainly focus on splitting the boundary of each element in a single document page, neglecting the reconstruction of semantic structure in multi-page documents. This paper introduces hierarchical reconstruction of document structures as a novel task suitable for NLP and CV fields. To better evaluate the system performance on the new task, we built a large-scale dataset named HRDoc, which consists of 2,500 multi-page documents with nearly 2 million semantic units. Every document in HRDoc has line-level annotations including categories and relations obtained from rule-based extractors and human annotators. Moreover, we proposed an encoder-decoder-based hierarchical document structure parsing system (DSPS) to tackle this problem. By adopting a multi-modal bidirectional encoder and a structure-aware GRU decoder with soft-mask operation, the DSPS model surpass the baseline method by a large margin. All scripts and datasets will be made publicly available at https://github.com/jfma-USTC/HRDoc.

Addressing the issues of who saying what to whom in multi-party conversations (MPCs) has recently attracted a lot of research attention.However, existing methods on MPC understanding typically embed interlocutors and utterances into sequential information flows, or utilize only the superficial of inherent graph structures in MPCs.To this end, we present a plug-and-play and lightweight method named graph-induced fine-tuning (GIFT) which can adapt various Transformer-based pre-trained language models (PLMs) for universal MPC understanding.In detail, the full and equivalent connections among utterances in regular Transformer ignore the sparse but distinctive dependency of an utterance on another in MPCs.To distinguish different relationships between utterances, four types of edges are designed to integrate graph-induced signals into attention mechanisms to refine PLMs originally designed for processing sequential texts.We evaluate GIFT by implementing it into three PLMs, and test the performance on three downstream tasks including addressee recognition, speaker identification and response selection.Experimental results show that GIFT can significantly improve the performance of three PLMs on three downstream tasks and two benchmarks with only 4 additional parameters per encoding layer, achieving new state-of-theart performance on MPC understanding.

LiDAR and Radar are two complementary sensing approaches in that LiDAR specializes in capturing an object's 3D shape while Radar provides longer detection ranges as well as velocity hints. Though seemingly natural, how to efficiently combine them for improved feature representation is still unclear. The main challenge arises from that Radar data are extremely sparse and lack height information. Therefore, directly integrating Radar features into LiDAR-centric detection networks is not optimal. In this work, we introduce a bi-directional LiDAR-Radar fusion framework, termed Bi-LRFusion, to tackle the challenges and improve 3D detection for dynamic objects. Technically, Bi-LRFusion involves two steps: first, it enriches Radar's local features by learning important details from the LiDAR branch to alleviate the problems caused by the absence of height information and extreme sparsity; second, it combines LiDAR features with the enhanced Radar features in a unified bird's-eye-view representation. We conduct extensive experiments on nuScenes and ORR datasets, and show that our Bi-LRFusion achieves state-of-the-art performance for detecting dynamic objects. Notably, Radar data in these two datasets have different formats, which demonstrates the generalizability of our method. Codes will be published.

Modeling multi-party conversations (MPCs) with graph neural networks has been proven effective at capturing complicated and graphical information flows.However, existing methods rely heavily on the necessary addressee labels and can only be applied to an ideal setting where each utterance must be tagged with an "@" or other equivalent addressee label.To study the scarcity of addressee labels which is a common issue in MPCs, we propose MADNet that maximizes addressee deduction expectation in heterogeneous graph neural networks for MPC generation.Given an MPC with a few addressee labels missing, existing methods fail to build a consecutively connected conversation graph, but only a few separate conversation fragments instead.To ensure message passing between these conversation fragments, four additional types of latent edges are designed to complete a fully-connected graph.Besides, to optimize the edge-typedependent message passing for those utterances without addressee labels, an Expectation-Maximization-based method that iteratively generates silver addressee labels (E step), and optimizes the quality of generated responses (M step), is designed.Experimental results on two Ubuntu IRC channel benchmarks show that MADNet outperforms various baseline models on the task of MPC generation, especially under the more common and challenging setting where part of addressee labels are missing.

The Multimodal Information based Speech Processing (MISP) 2022 challenge aimed to enhance speech processing performance in harsh acoustic environments by leveraging additional modalities such as video or text. The challenge included two tracks: audio-visual speaker diarization (AVSD) and audio-visual diarization and recognition (AVDR). The training material was based on previous MISP 2021 recordings, but we have accurately synchronized audio and visual data. Additionally, a new evaluation set was provided. This paper gives an overview of the challenge setup, presents the results, and summarizes the effective techniques employed by the participants. We also analyze the current technical challenges and suggest directions for future research in AVSD and AVDR.

The Multi-modal Information based Speech Processing (MISP) challenge aims to extend the application of signal processing technology in specific scenarios by promoting the research into wake-up words, speaker diarization, speech recognition, and other technologies. The MISP2022 challenge has two tracks: 1) audio-visual speaker diarization (AVSD), aiming to solve "who spoken when" using both audio and visual data; 2) a novel audio-visual diarization and recognition (AVDR) task that focuses on addressing "who spoken what when" with audio-visual speaker diarization results. Both tracks focus on the Chinese language, and use far-field audio and video in real home-tv scenarios: 2-6 people communicating each other with TV noise in the background. This paper introduces the dataset, track settings, and baselines of the MISP2022 challenge. Our analyses of experiments and examples indicate the good performance of AVDR baseline system, and the potential difficulties in this challenge due to, e.g., the far-field video quality, the presence of TV noise in the background, and the indistinguishable speakers.

In this work, we present a novel method, named AV2vec, for learning audio-visual speech representations by multimodal self-distillation. AV2vec has a student and a teacher module, in which the student performs a masked latent feature regression task using the multimodal target features generated online by the teacher. The parameters of the teacher model are a momentum update of the student. Since our target features are generated online, AV2vec needs no iteration step like AV-HuBERT and the total training time cost is reduced to less than one-fifth. We further propose AV2vec-MLM in this study, which augments AV2vec with a masked language model (MLM)-style loss using multitask learning. Our experimental results show that AV2vec achieved comparable performance to the AV-HuBERT baseline. When combined with an MLM-style loss, AV2vec-MLM outperformed baselines and achieved the best performance on the downstream tasks.

Recent audio2mesh-based methods have shown promising prospects for speech-driven 3D facial animation tasks. However, some intractable challenges are urgent to be settled. For example, the data-scarcity problem is intrinsically inevitable due to the difficulty of 4D data collection. Besides, current methods generally lack controllability on the animated face. To this end, we propose a novel framework named Speech4Mesh to consecutively generate 4D talking head data and train the audio2mesh network with the reconstructed meshes. In our framework, we first reconstruct the 4D talking head sequence based on the monocular videos. For precise capture of the talking-related variation on the face, we exploit the audio-visual alignment information from the video by employing a contrastive learning scheme. We next can train the audio2mesh network (e.g., FaceFormer) based on the generated 4D data. To get control of the animated talking face, we encode the speaking-unrelated factors (e.g., emotion, etc.) into an emotion embedding for manipulation. Finally, a differentiable renderer guarantees more accurate photometric details of the reconstruction and animation results. Empirical experiments demonstrate that the Speech4Mesh framework can not only outperform state-of-the-art reconstruction methods, especially on the lower-face part but also achieve better animation performance both perceptually and objectively after pre-trained on the synthesized data. Besides, we also verify that the proposed framework is able to explicitly control the emotion of the animated talking face.

Tremendous efforts have been made on document image rectification, but how to learn effective representation of such distorted images is still under-explored. In this paper, we present DocMAE, a novel self-supervised framework for document image rectification. Our motivation is to encode the structural cues in document images by leveraging masked autoencoder to benefit the rectification, i.e., the document boundaries, and text lines. Specifically, we first mask random patches of the background-excluded document images and then reconstruct the missing pixels. With such a self-supervised learning approach, the network is encouraged to learn the intrinsic structure of deformed documents by restoring document boundaries and missing text lines. Transfer performance in the downstream rectification task validates the effectiveness of our method. Extensive experiments are conducted to demonstrate the effectiveness of our method.

Although pre-trained language models~(PLMs) have recently advanced the research progress in mathematical reasoning, they are not specially designed as a capable multi-task solver, suffering from high cost for multi-task deployment (e.g. a model copy for a task) and inferior performance on complex mathematical problems in practical applications. To address these issues, we propose JiuZhang 2.0, a unified Chinese PLM specially for multi-task mathematical problem solving. Our idea is to maintain a moderate-sized model and employ the cross-task knowledge sharing to improve the model capacity in a multi-task setting. Specially, we construct a Mixture-of-Experts (MoE) architecture for modeling mathematical text, to capture the common mathematical knowledge across tasks. For optimizing the MoE architecture, we design multi-task continual pre-training and multi-task fine-tuning strategies for multi-task adaptation. These training strategies can effectively decompose the knowledge from the task data and establish the cross-task sharing via expert networks. To further improve the general capacity of solving different complex tasks, we leverage large language models (LLMs) as complementary models to iteratively refine the generated solution by our PLM, via in-context learning. Extensive experiments have demonstrated the effectiveness of our model.

Satisfactory recognition performance has been achieved for simple and controllable printed molecular images. However, recognizing handwritten chemical structure images remains unresolved due to the inherent ambiguities in handwritten atoms and bonds, as well as the signifcant challenge of converting projected 2D molecular layouts into markup strings. Target to address these problems, this paper proposes an end-to-end framework for handwritten chemical structure images recognition, with novel structure-specific markup language (SSML) and random conditional guided decoder (RCGD). SSML alleviates ambiguity and complexity in Chemfig syntax by designing an innovative markup language to accurately depict molecular structures. Besides, we propose RCGD to address the issue of multiple path decoding of molecular structures, which is composed of conditional attention guidance, memory classification and path selection mechanisms. In order to fully confirm the effectiveness of the end-to-end method, a new database containing 50,000 handwritten chemical structure images (EDU-CHEMC) has been established. Experimental results demonstrate that compared to traditional SMILES sequences, our SSML can significantly reduces the semantic gap between chemical images and markup strings. It is worth noting that our method can also recognize invalid or non-existent organic molecular structures, making it highly applicable for tasks related to teaching evaluations in the fields of chemistry and biology education. The EDU-CHEMC will be released soon in https://github.com/iFLYTEK-CV/EDU-CHEMC.

Zero-shot cross-lingual named entity recognition (NER) aims at transferring knowledge from annotated and rich-resource data in source languages to unlabeled and lean-resource data in target languages.Existing mainstream methods based on the teacher-student distillation framework ignore the rich and complementary information lying in the intermediate layers of pre-trained language models, and domaininvariant information is easily lost during transfer.In this study, a mixture of short-channel distillers (MSD) method is proposed to fully interact the rich hierarchical information in the teacher model and to transfer knowledge to the student model sufficiently and efficiently.Concretely, a multi-channel distillation framework is designed for sufficient information transfer by aggregating multiple distillers as a mixture.Besides, an unsupervised method adopting parallel domain adaptation is proposed to shorten the channels between the teacher and student models to preserve domaininvariant features.Experiments on four datasets across nine languages demonstrate that the proposed method achieves new state-of-the-art performance on zero-shot cross-lingual NER and shows great generalization and compatibility across languages and fields.

In this paper we discuss the rational of the Multi-model Information based Speech Processing (MISP) Challenge, and provide a detailed description of the data recorded, the two evaluation tasks and the corresponding baselines, followed by a summary of submitted systems and evaluation results. The MISP Challenge aims at tack-ling speech processing tasks in different scenarios by introducing information about an additional modality (e.g., video, or text), which will hopefully lead to better environmental and speaker robustness in realistic applications. In the first MISP challenge, two bench-mark datasets recorded in a real-home TV room with two reproducible open-source baseline systems have been released to promote research in audio-visual wake word spotting (AVWWS) and audio-visual speech recognition (AVSR). To our knowledge, MISP is the first open evaluation challenge to tackle real-world issues of AVWWS and AVSR in the home TV scenario.

This paper aims to advance the mathematical intelligence of machines by presenting the first Chinese mathematical pre-trained language model (PLM) for effectively understanding and representing mathematical problems. Unlike other standard NLP tasks, mathematical texts are difficult to understand, since they involve mathematical terminology, symbols and formulas in the problem statement. Typically, it requires complex mathematical logic and background knowledge for solving mathematical problems.

X-ray prohibited items detection of security inspection plays an important role in protecting public safety. It is a typical few-shot object detection (FSOD) task because some categories of prohibited items are highly scarce due to low-frequency appearance, e.g. pistols, which has been ignored by recent X-ray detection works. In contrast to most FSOD studies that rely on rich feature correlations from natural scenarios, the more practical X-ray security inspection usually faces the dilemma of only weak features learnable due to heavy occlusion, color fading, etc, which causes a severe performance drop when traditional FSOD methods are adopted. However, professional X-ray FSOD evaluation benchmarks and effective models of this scenario have been rarely studied in recent years. Therefore, in this paper, we propose the first X-ray FSOD dataset on the typical industrial X-ray security inspection scenario consisting of 12,333 images and 41,704 instances from 20 categories, which could benchmark and promote FSOD studies in such more challenging scenarios. Further, we propose the Weak-feature Enhancement Network (WEN) containing two core modules, i.e. Prototype Perception (PR) and Feature Reconciliation (FR), where PR first generates a prototype library by aggregating and extracting the basis feature from critical regions around instances, to generate the basis information for each category; FR then adaptively adjusts the impact intensity of the corresponding prototype and forces the model to precisely enhance the weak features of specific objects through the basis information. This mechanism is also effective in traditional FSOD tasks. Extensive experiments on X-ray FSOD and Pascal VOC datasets demonstrate that WEN outperforms other baselines in both X-ray and common scenarios.

Abstract Chest X‐ray images taken under different conditions follow different distributions, preventing the models trained on a domain from generalising well on the other domain. In this paper, a generative domain adaptation (GDA) method is proposed to address this issue and facilitate the learning process for downstream analysis. GDA adapts different domains to a virtual common one where images are aligned at the appearance level. To this end, a domain shared generator is used to transform the input images and two competitive discriminators are used to adversarially supervise the transforming process. The domain discriminator drives the generator to narrow the domain gap while the fidelity discriminator forces the generator to keep the inherent information. Moreover, a specific classification or detection network is attached to the generator to supervise it in a task‐oriented manner. Experiment results on a large‐scale dataset containing 46k chest X‐ray images demonstrate that GDA outperforms representative domain adaptation methods by a large margin for both disease classification and lesion detection as well as provides useful transformed images to assist experts for diagnosis.

A common strategy adopted by existing state-of-the-art unsupervised domain adaptation (UDA) methods is to employ two classifiers to identify the misaligned local regions between source and target domain. Following the 'wisdom of the crowd' principle, one has to ask: why stop at two? Indeed, we find that using more classifiers leads to better performance, but also introduces more model parameters, therefore risking overfitting. In this paper, we introduce a novel method called STochastic clAssifieRs (STAR) for addressing this problem. Instead of representing one classifier as a weight vector, STAR models it as a Gaussian distribution with its variance representing the inter-classifier discrepancy. With STAR, we can now sample an arbitrary number of classifiers from the distribution, whilst keeping the model size the same as having two classifiers. Extensive experiments demonstrate that a variety of existing UDA methods can greatly benefit from STAR and achieve the state-of-the-art performance on both image classification and semantic segmentation tasks.

Domain generalization (DG) is the challenging and topical problem of learning models that generalize to novel testing domains with different statistics than a set of known training domains. The simple approach of aggregating data from all source domains and training a single deep neural network end-to-end on all the data provides a surprisingly strong baseline that surpasses many prior published methods. In this paper we build on this strong baseline by designing an episodic training procedure that trains a single deep network in a way that exposes it to the domain shift that characterises a novel domain at runtime. Specifically, we decompose a deep network into feature extractor and classifier components, and then train each component by simulating it interacting with a partner who is badly tuned for the current domain. This makes both components more robust, ultimately leading to our networks producing state-of-the-art performance on three DG benchmarks. Furthermore, we consider the pervasive workflow of using an ImageNet trained CNN as a fixed feature extractor for downstream recognition tasks. Using the Visual Decathlon benchmark, we demonstrate that our episodic-DG training improves the performance of such a general purpose feature extractor by explicitly training a feature for robustness to novel problems. This shows that DG training can benefit standard practice in computer vision.

In this paper, we propose a novel neural network structure, namely feedforward sequential memory networks (FSMN), to model long-term dependence in time series without using recurrent feedback. The proposed FSMN is a standard fully connected feedforward neural network equipped with some learnable memory blocks in its hidden layers. The memory blocks use a tapped-delay line structure to encode the long context information into a fixed-size representation as short-term memory mechanism which are somehow similar to the time-delay neural networks layers. We have evaluated the FSMNs in several standard benchmark tasks, including speech recognition and language modeling. Experimental results have shown that FSMNs outperform the conventional recurrent neural networks (RNN) while can be learned much more reliably and faster in modeling sequential signals like speech or language. Moreover, we also propose a compact feedforward sequential memory networks (cFSMN) by combining FSMN with low-rank matrix factorization and make a slight modification to the encoding method used in FSMNs in order to further simplify the network architecture. On the speech recognition Switchboard task, the proposed cFSMN structures can reduce the model size by 60% and speed up the learning by more than seven times while the model can still significantly outperform the popular bidirectional LSTMs for both frame-level cross-entropy criterion-based training and MMI-based sequence training.

This paper presents a study on large vocabulary continuous whisper automatic recognition (wLVCSR). wLVCSR provides the ability to use ASR equipment in public places without concern for disturbing others or leaking private information. However the task of wLVCSR is much more challenging than normal LVCSR due to the absence of pitch which not only causes the signal to noise ratio (SNR) of whispers to be much lower than normal speech but also leads to flatness and formant shifts in whisper spectra. Furthermore, the amount of whisper data available for training is much less than for normal speech. In this paper, multi-task deep neural network (DNN) acoustic models are deployed to solve these problems. Moreover, model adaptation is performed on the multi-task DNN to normalize speaker and environmental variability in whispers based on discriminative speaker identity information. On a Mandarin whisper dictation task, with 55 hours of whisper data, the proposed SI multi-task DNN model can achieve 56.7% character error rate (CER) improvement over a baseline Gaussian Mixture Model (GMM), discriminatively trained only using the whisper data. Besides, the CER of the proposed model for normal speech can reach 15.2%, which is close to the performance of a state-of-the-art DNN trained with one thousand hours of speech data. From this baseline, the model-adapted DNN gains a further 10.9% CER reduction over the generic model.

Recently, the hybrid model combining deep neural network (DNN) with context-dependent HMMs has achieved some dramatic gains over the conventional GMM/HMM method in many speech recognition tasks. In this paper, we study how to compete with the state-of-the-art DNN/HMM method under the traditional GMM/HMM framework. Instead of using DNN as acoustic model, we use DNN as a front-end bottleneck (BN) feature extraction method to decorrelate long feature vectors concatenated from several consecutive speech frames. More importantly, we have proposed two novel incoherent training methods to explicitly de-correlate BN features in learning of DNN. The first method relies on minimizing coherence of weight matrices in DNN while the second one attempts to minimize correlation coefficients of BN features calculated in each mini-batch data in DNN training. Experimental results on a 70-hr Mandarin transcription task and the 309-hr Switchboard task have shown that the traditional GMM/HMMs using BN features can yield comparable performance as DNN/HMM. The proposed incoherent training can produce 2-3% additional gain over the baseline BN features. At last, the discriminatively trained GMM/HMMs using incoherently trained BN features have consistently surpassed the state-of-the-art DNN/HMMs in all evaluated tasks.

Recently a pre-trained context-dependent hybrid deep neural network (DNN) and HMM method has achieved significant performance gain in many large-scale automatic speech recognition (ASR) tasks. However, the error back-propagation (BP) algorithm for training neural networks is sequential in nature and is hard to parallelize into multiple computing threads. Therefore, training a deep neural network is extremely time-consuming even with a modern GPU board. In this paper we have proposed a new acoustic modelling framework to use multiple DNNs instead of a single DNN to compute the posterior probabilities of tied HMM states. In our method, all tied states of context-dependent HMMs are first grouped into several disjoined clusters based on the training data associated with these HMM states. Then, several hierarchically structured DNNs are trained separately for these disjoined clusters of data using multiple GPUs. In decoding, the final posterior probability of each tied HMM state can be calculated based on output posteriors from multiple DNNs. We have evaluated the proposed method on a 64-hour Mandarin transcription task and 309-hour Switchboard Hub5 task. Experimental results have shown that the new method using clusterbased multiple DNNs can achieve over 5 times reduction in total training time with only negligible performance degradation (about 1-2% in average) when using 3 or 4 GPUs respectively.

In this paper, we have proposed two novel optimization methods for discriminative training (DT) of hidden Markov models (HMMs) in speech recognition based on an efficient global optimization algorithm used to solve the so-called trust region (TR) problem, where a quadratic function is minimized under a spherical constraint. In the first method, maximum mutual information estimation (MMIE) of Gaussian mixture HMMs is formulated as a standard TR problem so that the efficient global optimization method can be used in each iteration to maximize the auxiliary function of discriminative training for speech recognition. In the second method, we propose to construct a new auxiliary function for DT of HMMs by adding a quadratic penalty term. The new auxiliary function is constructed to serve as first-order approximation as well as lower bound of the original discriminative objective function within a locality constraint. Due to the lower-bound property, the found optimal point of the new auxiliary function is guaranteed to improve the original discriminative objective function until it converges to a local optimum or stationary point of the objective function. Both TR-based optimization methods have been investigated on two standard large-vocabulary continuous speech recognition tasks, using the WSJ0 and Switchboard databases. Experimental results have shown that the proposed TR methods outperform the conventional EBW method in terms of convergence behavior as well as recognition performance.

In this paper, we have proposed a new method to construct an auxiliary function for the discriminative training of HMMs in speech recognition. The new auxiliary function serves as a first-order approximation of the original objective function but more importantly it remains as a lower bound of the original objective function as well. Furthermore, the trust region (TR) method in [1] is applied to find the globally optimal point of the new auxiliary function. Due to its lower-bound property, the found optimal point is theoretically guaranteed to increase the original discriminative objective function. The proposed bounded trust region method has been investigated on two LVCSR tasks, namely WSJ-5k and Switchboard 60-hour subset tasks. Experimental results show that the bounded TR method yields much better convergence behavior than both the conventional EBW method and the original TR method.

In this paper, we present a new optimization method for MMIE-based discriminative training of HMMs in speech recognition. In our method, the MMIE training of Gaussian mixture HMMs is formulated as a so-called trust region problem, where a quadratic objective function is minimized under a spherical constraint, so that an efficient global optimization method for the trust region problem can be used to solve the MMIE training problem of HMMs. Experimental results on the WSJ0 Nov'92 evaluation task demonstrate that the trust region based optimization significantly outperforms the conventional EBWmethod in terms of optimization convergence behavior as well as speech recognition performance. It has been observed that the trust region method achieves up to 23.3% relative recognition error reduction over a well-trained MLE system while the EBW method gives only 13.3% relative error reduction.

In this paper, we propose a novel optimization algorithm called constrained line search (CLS) for discriminative training (DT) of Gaussian mixture continuous density hidden Markov model (CDHMM) in speech recognition. The CLS method is formulated under a general framework for optimizing any discriminative objective functions including maximum mutual information (MMI), minimum classification error (MCE), minimum phone error (MPE)/minimum word error (MWE), etc. In this method, discriminative training of HMM is first cast as a constrained optimization problem, where Kullback-Leibler divergence (KLD) between models is explicitly imposed as a constraint during optimization. Based upon the idea of line search, we show that a simple formula of HMM parameters can be found by constraining the KLD between HMM of two successive iterations in an quadratic form. The proposed CLS method can be applied to optimize all model parameters in Gaussian mixture CDHMMs, including means, covariances, and mixture weights. We have investigated the proposed CLS approach on several benchmark speech recognition databases, including TIDIGITS, Resource Management (RM), and Switchboard. Experimental results show that the new CLS optimization method consistently outperforms the conventional EBW method in both recognition performance and convergence behavior.

Recently, we proposed a novel optimization algorithm called constrained line search (CLS) to train Gaussian mean vectors of HMMs in the MMI sense. In this paper, we extend and re-formulate it in a more general framework. The new CLS can optimize any discriminative objective functions including MMI, MCE, MPE/MWE etc. Also, closed-form solutions to update all Gaussian mixture parameters, including means, covariances and mixture weights, are obtained. We investigate the new CLS on several benchmark speech recognition databases, including TIDIGITS, Switchboard mini-train and Switchboard full h5train00 sets. Experimental results show that the new CLS optimization method outperforms the conventional EBW method in both performance and convergence behavior.

In this paper, we propose a novel constrained line search to optimize the MMEE objective function for training discriminative HMMs. In our method, the MMI estimation is cast as a constrained maximization problem, where Kullback-Leibler divergence between models before and after parameters adjustment is introduced as a constraint during optimization. Then, based on the idea of line search, we show that a simple, closed-form solution can be derived under some approximation assumptions. The proposed optimization method have been investigated in two speech recognition tasks: TIDIGITS and Switchboard (mini-train). Experimental results show that the new training method achieves significant word error rate reduction when comparing with our best MLE models, i.e., relatively 63.8% on TIDIGITS and 6.1% on the Switchboard mini-train set, respectively. Our results also show that the constrained line search method consistently outperforms the popular EBW method in both tasks.