Xiangteng He

Peng Wu, Jing Liu, Xiangteng He et al.

Video anomaly detection (VAD) has been paid increasing attention due to its potential applications, its current dominant tasks focus on online detecting anomalies% at the frame level, which can be roughly interpreted as the binary or multiple event classification. However, such a setup that builds relationships between complicated anomalous events and single labels, e.g., ``vandalism'', is superficial, since single labels are deficient to characterize anomalous events. In reality, users tend to search a specific video rather than a series of approximate videos. Therefore, retrieving anomalous events using detailed descriptions is practical and positive but few researches focus on this. In this context, we propose a novel task called Video Anomaly Retrieval (VAR), which aims to pragmatically retrieve relevant anomalous videos by cross-modalities, e.g., language descriptions and synchronous audios. Unlike the current video retrieval where videos are assumed to be temporally well-trimmed with short duration, VAR is devised to retrieve long untrimmed videos which may be partially relevant to the given query. To achieve this, we present two large-scale VAR benchmarks, UCFCrime-AR and XDViolence-AR, constructed on top of prevalent anomaly datasets. Meanwhile, we design a model called Anomaly-Led Alignment Network (ALAN) for VAR. In ALAN, we propose an anomaly-led sampling to focus on key segments in long untrimmed videos. Then, we introduce an efficient pretext task to enhance semantic associations between video-text fine-grained representations. Besides, we leverage two complementary alignments to further match cross-modal contents. Experimental results on two benchmarks reveal the challenges of VAR task and also demonstrate the advantages of our tailored method. Captions are publicly released at https://github.com/Roc-Ng/VAR.

Hongbo Sun, Xiangteng He, Yuxin Peng

Fine-grained visual categorization (FGVC) aims at recognizing objects from similar subordinate categories, which is challenging and practical for human's accurate automatic recognition needs. Most FGVC approaches focus on the attention mechanism research for discriminative regions mining while neglecting their interdependencies and composed holistic object structure, which are essential for model's discriminative information localization and understanding ability. To address the above limitations, we propose the Structure Information Modeling Transformer (SIM-Trans) to incorporate object structure information into transformer for enhancing discriminative representation learning to contain both the appearance information and structure information. Specifically, we encode the image into a sequence of patch tokens and build a strong vision transformer framework with two well-designed modules: (i) the structure information learning (SIL) module is proposed to mine the spatial context relation of significant patches within the object extent with the help of the transformer's self-attention weights, which is further injected into the model for importing structure information; (ii) the multi-level feature boosting (MFB) module is introduced to exploit the complementary of multi-level features and contrastive learning among classes to enhance feature robustness for accurate recognition. The proposed two modules are light-weighted and can be plugged into any transformer network and trained end-to-end easily, which only depends on the attention weights that come with the vision transformer itself. Extensive experiments and analyses demonstrate that the proposed SIM-Trans achieves state-of-the-art performance on fine-grained visual categorization benchmarks. The code is available at https://github.com/PKU-ICST-MIPL/SIM-Trans_ACMMM2022.

Yulin Pan, Xiangteng He, Biao Gong et al.

Video temporal grounding aims to pinpoint a video segment that matches the query description. Despite the recent advance in short-form videos (\textit{e.g.}, in minutes), temporal grounding in long videos (\textit{e.g.}, in hours) is still at its early stage. To address this challenge, a common practice is to employ a sliding window, yet can be inefficient and inflexible due to the limited number of frames within the window. In this work, we propose an end-to-end framework for fast temporal grounding, which is able to model an hours-long video with \textbf{one-time} network execution. Our pipeline is formulated in a coarse-to-fine manner, where we first extract context knowledge from non-overlapped video clips (\textit{i.e.}, anchors), and then supplement the anchors that highly response to the query with detailed content knowledge. Besides the remarkably high pipeline efficiency, another advantage of our approach is the capability of capturing long-range temporal correlation, thanks to modeling the entire video as a whole, and hence facilitates more accurate grounding. Experimental results suggest that, on the long-form video datasets MAD and Ego4d, our method significantly outperforms state-of-the-arts, and achieves \textbf{14.6$\times$} / \textbf{102.8$\times$} higher efficiency respectively. Project can be found at \url{https://github.com/afcedf/SOONet.git}.

HsiaoYuan Hsu, Xiangteng He, Yuxin Peng et al.

Content-aware visual-textual presentation layout aims at arranging spatial space on the given canvas for pre-defined elements, including text, logo, and underlay, which is a key to automatic template-free creative graphic design. In practical applications, e.g., poster designs, the canvas is originally non-empty, and both inter-element relationships as well as inter-layer relationships should be concerned when generating a proper layout. A few recent works deal with them simultaneously, but they still suffer from poor graphic performance, such as a lack of layout variety or spatial non-alignment. Since content-aware visual-textual presentation layout is a novel task, we first construct a new dataset named PosterLayout, which consists of 9,974 poster-layout pairs and 905 images, i.e., non-empty canvases. It is more challenging and useful for greater layout variety, domain diversity, and content diversity. Then, we propose design sequence formation (DSF) that reorganizes elements in layouts to imitate the design processes of human designers, and a novel CNN-LSTM-based conditional generative adversarial network (GAN) is presented to generate proper layouts. Specifically, the discriminator is design-sequence-aware and will supervise the "design" process of the generator. Experimental results verify the usefulness of the new benchmark and the effectiveness of the proposed approach, which achieves the best performance by generating suitable layouts for diverse canvases.

Ailar Mahdizadeh, Puria Azadi, Muchen Li et al.

Streaming video understanding with large vision-language models (VLMs) requires a compact memory that can support future reasoning over an ever-growing visual history. A common solution is to compress the key-value (KV) cache, but existing streaming methods typically rely on local token-wise heuristics, such as recency, temporal redundancy, or saliency, which do not explicitly optimize whether the retained cache is representative of the accumulated history. We propose to view KV-cache compression as a coreset selection problem: rather than scoring tokens independently for retention, we select a small subset that covers the geometry of the accumulated visual cache. Our method operates in a joint KV representation and introduces a bicriteria objective that balances coverage in key and value spaces, preserving both retrieval structure and output-relevant information. To encourage a more diverse retained subset, we further introduce an orthogonality-driven diversity criterion that favors candidates contributing new directions beyond the current selection, and connect this criterion to log-determinant subset selection. Across four open-source VLMs and five long-video and streaming-video benchmarks, our method improves over heuristic streaming compression baselines under a fixed cache budget. These results highlight that representative coreset selection offers a more effective principle, than token-wise pruning, for memory-constrained streaming video understanding.

Ailar Mahdizadeh, Puria Azadi Moghadam, Xiangteng He et al.

Vision-language models (VLMs) have demonstrated strong cross-modal capabilities, yet most work remains limited to 2D data and assumes binary supervision (i.e., positive vs. negative pairs), overlooking the continuous and structured dependencies present in volumetric data such as CT. Existing approaches often treat volumetric scans as independent 2D slices, compromising spatial coherence and underutilizing rich clinical semantics. We propose SCALE-VLP, a soft-weighted contrastive vision-language pre-training framework that integrates (i) volumetric spatial semantics to preserve anatomical structure and (ii) domain-aware, knowledge-infused semantics (e.g., radiological ontologies) to guide alignment. This yields structurally consistent and semantically grounded representations under limited supervision, demonstrating strong cross-task transferability (retrieval, report generation, and classification), and cross-domain generalizability with consistent gains without further fine-tuning. In particular, compared to the previous state of the art, SCALE-VLP achieves up to 4.3x higher top-1 CT-report retrieval, improves abnormality classification by 10 points, and reaches ROUGE-L 0.44 and BERT-F1 0.89 for report generation. Further, in zero-shot evaluation on an out-of-domain external dataset, we observe consistent gains, indicating the cross-task and cross-domain generalization ability of SCALE-VLP.

Kun Yuan, Min Woo Sun, Zhen Chen et al.

There is a growing interest in developing strong biomedical vision-language models. A popular approach to achieve robust representations is to use web-scale scientific data. However, current biomedical vision-language pretraining typically compresses rich scientific figures and text into coarse figure-level pairs, discarding the fine-grained correspondences that clinicians actually rely on when zooming into local structures. To tackle this issue, we introduce Panel2Patch, a novel data pipeline that mines hierarchical structure from existing biomedical scientific literature, i.e., multi-panel, marker-heavy figures and their surrounding text, and converts them into multi-granular supervision. Given scientific figures and captions, Panel2Patch parses layouts, panels, and visual markers, then constructs hierarchical aligned vision-language pairs at the figure, panel, and patch levels, preserving local semantics instead of treating each figure as a single data sample. Built on this hierarchical corpus, we develop a granularity-aware pretraining strategy that unifies heterogeneous objectives from coarse didactic descriptions to fine region-focused phrases. By applying Panel2Patch to only a small set of the literature figures, we extract far more effective supervision than prior pipelines, enabling substantially better performance with less pretraining data.

Yulin Pan, Xiangteng He, Chaojie Mao et al.

Image generation has witnessed significant advancements in the past few years. However, evaluating the performance of image generation models remains a formidable challenge. In this paper, we propose ICE-Bench, a unified and comprehensive benchmark designed to rigorously assess image generation models. Its comprehensiveness could be summarized in the following key features: (1) Coarse-to-Fine Tasks: We systematically deconstruct image generation into four task categories: No-ref/Ref Image Creating/Editing, based on the presence or absence of source images and reference images. And further decompose them into 31 fine-grained tasks covering a broad spectrum of image generation requirements, culminating in a comprehensive benchmark. (2) Multi-dimensional Metrics: The evaluation framework assesses image generation capabilities across 6 dimensions: aesthetic quality, imaging quality, prompt following, source consistency, reference consistency, and controllability. 11 metrics are introduced to support the multi-dimensional evaluation. Notably, we introduce VLLM-QA, an innovative metric designed to assess the success of image editing by leveraging large models. (3) Hybrid Data: The data comes from real scenes and virtual generation, which effectively improves data diversity and alleviates the bias problem in model evaluation. Through ICE-Bench, we conduct a thorough analysis of existing generation models, revealing both the challenging nature of our benchmark and the gap between current model capabilities and real-world generation requirements. To foster further advancements in the field, we will open-source ICE-Bench, including its dataset, evaluation code, and models, thereby providing a valuable resource for the research community.

Xiangteng He, Shunsuke Sakai, Kun Yuan et al.

Image-based Joint-Embedding Predictive Architecture (I-JEPA) learns visual representations by predicting latent embeddings of masked regions from visible context. However, it treats all regions uniformly and independently, lacking an explicit notion of where or in what order predictions should be made. Inspired by human visual perception, which deploys attention selectively and sequentially from the most informative to secondary regions, we propose DSeq-JEPA, a Discriminative Sequential Joint-Embedding Predictive Architecture that bridges predictive and autoregressive self-supervised learning, integrating JEPA-style latent prediction with GPT-style sequential reasoning. Specifically, DSeq-JEPA (i) first identifies primary discriminative regions based on a transformer-derived saliency map, emphasizing the distribution of visual importance, and then (ii) predicts subsequent regions in this discriminative order, progressively forming a curriculum-like semantic progression from primary to secondary cues -- a form of GPT-style pre-training. Extensive experiments across diverse tasks, including image classification (ImageNet), fine-grained visual categorization (iNaturalist21, CUB-200-2011, Stanford-Cars), detection and segmentation (MS-COCO, ADE20K), and low-level reasoning tasks (Clevr/Count, Clevr/Dist), demonstrate that DSeq-JEPA consistently focuses on more discriminative and generalizable representations than I-JEPA variants. Project page: https://github.com/SkyShunsuke/DSeq-JEPA.

Xiangteng He, Yuxin Peng, Liu Xie

Cross-media retrieval is to return the results of various media types corresponding to the query of any media type. Existing researches generally focus on coarse-grained cross-media retrieval. When users submit an image of "Slaty-backed Gull" as a query, coarse-grained cross-media retrieval treats it as "Bird", so that users can only get the results of "Bird", which may include other bird species with similar appearance (image and video), descriptions (text) or sounds (audio), such as "Herring Gull". Such coarse-grained cross-media retrieval is not consistent with human lifestyle, where we generally have the fine-grained requirement of returning the exactly relevant results of "Slaty-backed Gull" instead of "Herring Gull". However, few researches focus on fine-grained cross-media retrieval, which is a highly challenging and practical task. Therefore, in this paper, we first construct a new benchmark for fine-grained cross-media retrieval, which consists of 200 fine-grained subcategories of the "Bird", and contains 4 media types, including image, text, video and audio. To the best of our knowledge, it is the first benchmark with 4 media types for fine-grained cross-media retrieval. Then, we propose a uniform deep model, namely FGCrossNet, which simultaneously learns 4 types of media without discriminative treatments. We jointly consider three constraints for better common representation learning: classification constraint ensures the learning of discriminative features, center constraint ensures the compactness characteristic of the features of the same subcategory, and ranking constraint ensures the sparsity characteristic of the features of different subcategories. Extensive experiments verify the usefulness of the new benchmark and the effectiveness of our FGCrossNet. They will be made available at https://github.com/PKU-ICST-MIPL/FGCrossNet_ACMMM2019.

Yulin Pan, Chaojie Mao, Zeyinzi Jiang et al.

Prior studies have made significant progress in image inpainting guided by either text description or subject image. However, the research on inpainting with flexible guidance or control, i.e., text-only, image-only, and their combination, is still in the early stage. Therefore, in this paper, we introduce the multimodal promptable image inpainting project: a new task model, and data for taming customized image inpainting. We propose LAR-Gen, a novel approach for image inpainting that enables seamless inpainting of specific region in images corresponding to the mask prompt, incorporating both the text prompt and image prompt. Our LAR-Gen adopts a coarse-to-fine manner to ensure the context consistency of source image, subject identity consistency, local semantic consistency to the text description, and smoothness consistency. It consists of three mechanisms: (i) Locate mechanism: concatenating the noise with masked scene image to achieve precise regional editing, (ii) Assign mechanism: employing decoupled cross-attention mechanism to accommodate multi-modal guidance, and (iii) Refine mechanism: using a novel RefineNet to supplement subject details. Additionally, to address the issue of scarce training data, we introduce a novel data engine to automatically extract substantial pairs of data consisting of local text prompts and corresponding visual instances from a vast image data, leveraging publicly available pre-trained large models. Extensive experiments and various application scenarios demonstrate the superiority of LAR-Gen in terms of both identity preservation and text semantic consistency.

Jiayun Luo, Wan-Cyuan Fan, Lyuyang Wang et al.

Large Vision Language Models (LVLMs) have recently emerged as powerful architectures capable of understanding and reasoning over both visual and textual information. These models typically rely on two key components: a Vision Transformer (ViT) and a Large Language Model (LLM). ViT encodes visual content into a sequence of image tokens and serves as the perceptual front-end -- the eyes of the model. In contrast, the LLM interprets these tokens to perform high-level reasoning, generates responses, and functions as the cognitive core -- the brain of the model. However, it remains unclear which visual tokens contribute most significantly to understanding and reasoning, and how effectively these signals are propagated from ViT to the LLM. While most existing works have focused on identifying attention sinks, low-semantic tokens receiving disproportionately high attention, within the LLM, we shift the focus to the vision encoder by identifying a class of high-norm visual tokens from ViT, referred to as ViT attention sinks -- a problem that has been rarely studied but is indeed very important for LVLMs. Our findings show that these ViT sinks encapsulate high-level semantic concepts from images, allowing the LLM to perform more effective understanding and reasoning. Despite their importance, these sink tokens are often overlooked in existing LVLM architectures. To explore their contribution, we present both qualitative and quantitative analyses of the information embedded in these sink tokens. We also propose both training-free and training-based approaches to better leverage how this information is interpreted by the LLM, and to what extent. By explicitly utilizing these tokens, we demonstrate substantial improvements across a range of LVLMs and visual reasoning tasks, highlighting the untapped potential of ViT attention sinks in enhancing visual reasoning.

Shunsuke Sakai, Xiangteng He, Chunzhi Gu et al.

Despite the remarkable success, recent reconstruction-based anomaly detection (AD) methods via diffusion modeling still involve fine-grained noise-strength tuning and computationally expensive multi-step denoising, leading to a fundamental tension between fidelity and efficiency. In this paper, we propose a novel inversion-based AD approach - detection via noising in latent space - which circumvents explicit reconstruction. Importantly, we contend that the limitations in prior reconstruction-based methods originate from the prevailing detection via denoising in RGB space paradigm. To address this, we model AD under a reconstruction-free formulation, which directly infers the final latent variable corresponding to the input image via DDIM inversion, and then measures the deviation based on the known prior distribution for anomaly scoring. Specifically, in approximating the original probability flow ODE using the Euler method, we only enforce very few inversion steps to noise the clean image to pursue inference efficiency. As the added noise is adaptively derived with the learned diffusion model, the original features for the clean testing image can still be leveraged to yield high detection accuracy. We perform extensive experiments and detailed analysis across three widely used image AD datasets under the unsupervised unified setting to demonstrate the effectiveness of our model, regarding state-of-the-art AD performance, and about 2 times inference time speedup without diffusion distillation.

Zhen Han, Xiangteng He, Mingqian Tang et al.

Existing video copy detection methods generally measure video similarity based on spatial similarities between key frames, neglecting the latent similarity in temporal dimension, so that the video similarity is biased towards spatial information. There are methods modeling unified video similarity in an end-to-end way, but losing detailed partial alignment information, which causes the incapability of copy segments localization. To address the above issues, we propose the Video Similarity and Alignment Learning (VSAL) approach, which jointly models spatial similarity, temporal similarity and partial alignment. To mitigate the spatial similarity bias, we model the temporal similarity as the mask map predicted from frame-level spatial similarity, where each element indicates the probability of frame pair lying right on the partial alignments. To further localize partial copies, the step map is learned from the spatial similarity where the elements indicate extending directions of the current partial alignments on the spatial-temporal similarity map. Obtained from the mask map, the start points extend out into partial optimal alignments following instructions of the step map. With the similarity and alignment learning strategy, VSAL achieves the state-of-the-art F1-score on VCDB core dataset. Furthermore, we construct a new benchmark of partial video copy detection and localization by adding new segment-level annotations for FIVR-200k dataset, where VSAL also achieves the best performance, verifying its effectiveness in more challenging situations. Our project is publicly available at https://pvcd-vsal.github.io/vsal/.

Peng Wu, Xiangteng He, Mingqian Tang et al.

Video-text retrieval is an important yet challenging task in vision-language understanding, which aims to learn a joint embedding space where related video and text instances are close to each other. Most current works simply measure the video-text similarity based on video-level and text-level embeddings. However, the neglect of more fine-grained or local information causes the problem of insufficient representation. Some works exploit the local details by disentangling sentences, but overlook the corresponding videos, causing the asymmetry of video-text representation. To address the above limitations, we propose a Hierarchical Alignment Network (HANet) to align different level representations for video-text matching. Specifically, we first decompose video and text into three semantic levels, namely event (video and text), action (motion and verb), and entity (appearance and noun). Based on these, we naturally construct hierarchical representations in the individual-local-global manner, where the individual level focuses on the alignment between frame and word, local level focuses on the alignment between video clip and textual context, and global level focuses on the alignment between the whole video and text. Different level alignments capture fine-to-coarse correlations between video and text, as well as take the advantage of the complementary information among three semantic levels. Besides, our HANet is also richly interpretable by explicitly learning key semantic concepts. Extensive experiments on two public datasets, namely MSR-VTT and VATEX, show the proposed HANet outperforms other state-of-the-art methods, which demonstrates the effectiveness of hierarchical representation and alignment. Our code is publicly available.

Xiangteng He, Yulin Pan, Mingqian Tang et al.

Content-based video retrieval aims to find videos from a large video database that are similar to or even near-duplicate of a given query video. Video representation and similarity search algorithms are crucial to any video retrieval system. To derive effective video representation, most video retrieval systems require a large amount of manually annotated data for training, making it costly inefficient. In addition, most retrieval systems are based on frame-level features for video similarity searching, making it expensive both storage wise and search wise. We propose a novel video retrieval system, termed SVRTN, that effectively addresses the above shortcomings. It first applies self-supervised training to effectively learn video representation from unlabeled data to avoid the expensive cost of manual annotation. Then, it exploits transformer structure to aggregate frame-level features into clip-level to reduce both storage space and search complexity. It can learn the complementary and discriminative information from the interactions among clip frames, as well as acquire the frame permutation and missing invariant ability to support more flexible retrieval manners. Comprehensive experiments on two challenging video retrieval datasets, namely FIVR-200K and SVD, verify the effectiveness of our proposed SVRTN method, which achieves the best performance of video retrieval on accuracy and efficiency.

Xiangteng He, Yuxin Peng, Junjie Zhao

Fine-grained image classification is to recognize hundreds of subcategories in each basic-level category. Existing methods employ discriminative localization to find the key distinctions among subcategories. However, they generally have two limitations: (1) Discriminative localization relies on region proposal methods to hypothesize the locations of discriminative regions, which are time-consuming. (2) The training of discriminative localization depends on object or part annotations, which are heavily labor-consuming. It is highly challenging to address the two key limitations simultaneously, and existing methods only focus on one of them. Therefore, we propose a weakly supervised discriminative localization approach (WSDL) for fast fine-grained image classification to address the two limitations at the same time, and its main advantages are: (1) n-pathway end-to-end discriminative localization network is designed to improve classification speed, which simultaneously localizes multiple different discriminative regions for one image to boost classification accuracy, and shares full-image convolutional features generated by region proposal network to accelerate the process of generating region proposals as well as reduce the computation of convolutional operation. (2) Multi-level attention guided localization learning is proposed to localize discriminative regions with different focuses automatically, without using object and part annotations, avoiding the labor consumption. Different level attentions focus on different characteristics of the image, which are complementary and boost the classification accuracy. Both are jointly employed to simultaneously improve classification speed and eliminate dependence on object and part annotations. Compared with state-of-the-art methods on 2 widely-used fine-grained image classification datasets, our WSDL approach achieves the best performance.

Xiangteng He, Yuxin Peng, Junjie Zhao

Discriminative localization is essential for fine-grained image classification task, which devotes to recognizing hundreds of subcategories in the same basic-level category. Reflecting on discriminative regions of objects, key differences among different subcategories are subtle and local. Existing methods generally adopt a two-stage learning framework: The first stage is to localize the discriminative regions of objects, and the second is to encode the discriminative features for training classifiers. However, these methods generally have two limitations: (1) Separation of the two-stage learning is time-consuming. (2) Dependence on object and parts annotations for discriminative localization learning leads to heavily labor-consuming labeling. It is highly challenging to address these two important limitations simultaneously. Existing methods only focus on one of them. Therefore, this paper proposes the discriminative localization approach via saliency-guided Faster R-CNN to address the above two limitations at the same time, and our main novelties and advantages are: (1) End-to-end network based on Faster R-CNN is designed to simultaneously localize discriminative regions and encode discriminative features, which accelerates classification speed. (2) Saliency-guided localization learning is proposed to localize the discriminative region automatically, avoiding labor-consuming labeling. Both are jointly employed to simultaneously accelerate classification speed and eliminate dependence on object and parts annotations. Comparing with the state-of-the-art methods on the widely-used CUB-200-2011 dataset, our approach achieves both the best classification accuracy and efficiency.

Xiangteng He, Yuxin Peng

Fine-grained visual categorization is to recognize hundreds of subcategories belonging to the same basic-level category, which is a highly challenging task due to the quite subtle and local visual distinctions among similar subcategories. Most existing methods generally learn part detectors to discover discriminative regions for better categorization performance. However, not all parts are beneficial and indispensable for visual categorization, and the setting of part detector number heavily relies on prior knowledge as well as experimental validation. As is known to all, when we describe the object of an image via textual descriptions, we mainly focus on the pivotal characteristics, and rarely pay attention to common characteristics as well as the background areas. This is an involuntary transfer from human visual attention to textual attention, which leads to the fact that textual attention tells us how many and which parts are discriminative and significant to categorization. So textual attention could help us to discover visual attention in image. Inspired by this, we propose a fine-grained visual-textual representation learning (VTRL) approach, and its main contributions are: (1) Fine-grained visual-textual pattern mining devotes to discovering discriminative visual-textual pairwise information for boosting categorization performance through jointly modeling vision and text with generative adversarial networks (GANs), which automatically and adaptively discovers discriminative parts. (2) Visual-textual representation learning jointly combines visual and textual information, which preserves the intra-modality and inter-modality information to generate complementary fine-grained representation, as well as further improves categorization performance.

Xiangteng He, Yuxin Peng

Fine-grained image classification is a challenging task due to the large intra-class variance and small inter-class variance, aiming at recognizing hundreds of sub-categories belonging to the same basic-level category. Most existing fine-grained image classification methods generally learn part detection models to obtain the semantic parts for better classification accuracy. Despite achieving promising results, these methods mainly have two limitations: (1) not all the parts which obtained through the part detection models are beneficial and indispensable for classification, and (2) fine-grained image classification requires more detailed visual descriptions which could not be provided by the part locations or attribute annotations. For addressing the above two limitations, this paper proposes the two-stream model combining vision and language (CVL) for learning latent semantic representations. The vision stream learns deep representations from the original visual information via deep convolutional neural network. The language stream utilizes the natural language descriptions which could point out the discriminative parts or characteristics for each image, and provides a flexible and compact way of encoding the salient visual aspects for distinguishing sub-categories. Since the two streams are complementary, combining the two streams can further achieves better classification accuracy. Comparing with 12 state-of-the-art methods on the widely used CUB-200-2011 dataset for fine-grained image classification, the experimental results demonstrate our CVL approach achieves the best performance.

Yuxin Peng, Xiangteng He, Junjie Zhao

Fine-grained image classification is to recognize hundreds of subcategories belonging to the same basic-level category, such as 200 subcategories belonging to the bird, which is highly challenging due to large variance in the same subcategory and small variance among different subcategories. Existing methods generally first locate the objects or parts and then discriminate which subcategory the image belongs to. However, they mainly have two limitations: (1) Relying on object or part annotations which are heavily labor consuming. (2) Ignoring the spatial relationships between the object and its parts as well as among these parts, both of which are significantly helpful for finding discriminative parts. Therefore, this paper proposes the object-part attention model (OPAM) for weakly supervised fine-grained image classification, and the main novelties are: (1) Object-part attention model integrates two level attentions: object-level attention localizes objects of images, and part-level attention selects discriminative parts of object. Both are jointly employed to learn multi-view and multi-scale features to enhance their mutual promotions. (2) Object-part spatial constraint model combines two spatial constraints: object spatial constraint ensures selected parts highly representative, and part spatial constraint eliminates redundancy and enhances discrimination of selected parts. Both are jointly employed to exploit the subtle and local differences for distinguishing the subcategories. Importantly, neither object nor part annotations are used in our proposed approach, which avoids the heavy labor consumption of labeling. Comparing with more than 10 state-of-the-art methods on 4 widely-used datasets, our OPAM approach achieves the best performance.