Jun Long

Ruining Deng, Quan Liu, Can Cui et al.

Comprehensive semantic segmentation on renal pathological images is challenging due to the heterogeneous scales of the objects. For example, on a whole slide image (WSI), the cross-sectional areas of glomeruli can be 64 times larger than that of the peritubular capillaries, making it impractical to segment both objects on the same patch, at the same scale. To handle this scaling issue, prior studies have typically trained multiple segmentation networks in order to match the optimal pixel resolution of heterogeneous tissue types. This multi-network solution is resource-intensive and fails to model the spatial relationship between tissue types. In this paper, we propose the Omni-Seg+ network, a scale-aware dynamic neural network that achieves multi-object (six tissue types) and multi-scale (5X to 40X scale) pathological image segmentation via a single neural network. The contribution of this paper is three-fold: (1) a novel scale-aware controller is proposed to generalize the dynamic neural network from single-scale to multi-scale; (2) semi-supervised consistency regularization of pseudo-labels is introduced to model the inter-scale correlation of unannotated tissue types into a single end-to-end learning paradigm; and (3) superior scale-aware generalization is evidenced by directly applying a model trained on human kidney images to mouse kidney images, without retraining. By learning from ~150,000 human pathological image patches from six tissue types at three different resolutions, our approach achieved superior segmentation performance according to human visual assessment and evaluation of image-omics (i.e., spatial transcriptomics). The official implementation is available at https://github.com/ddrrnn123/Omni-Seg.

Xiangzuo Huo, Gang Sun, Shengwei Tian et al.

Medical image classification has developed rapidly under the impetus of the convolutional neural network (CNN). Due to the fixed size of the receptive field of the convolution kernel, it is difficult to capture the global features of medical images. Although the self-attention-based Transformer can model long-range dependencies, it has high computational complexity and lacks local inductive bias. Much research has demonstrated that global and local features are crucial for image classification. However, medical images have a lot of noisy, scattered features, intra-class variation, and inter-class similarities. This paper proposes a three-branch hierarchical multi-scale feature fusion network structure termed as HiFuse for medical image classification as a new method. It can fuse the advantages of Transformer and CNN from multi-scale hierarchies without destroying the respective modeling so as to improve the classification accuracy of various medical images. A parallel hierarchy of local and global feature blocks is designed to efficiently extract local features and global representations at various semantic scales, with the flexibility to model at different scales and linear computational complexity relevant to image size. Moreover, an adaptive hierarchical feature fusion block (HFF block) is designed to utilize the features obtained at different hierarchical levels comprehensively. The HFF block contains spatial attention, channel attention, residual inverted MLP, and shortcut to adaptively fuse semantic information between various scale features of each branch. The accuracy of our proposed model on the ISIC2018 dataset is 7.6% higher than baseline, 21.5% on the Covid-19 dataset, and 10.4% on the Kvasir dataset. Compared with other advanced models, the HiFuse model performs the best. Our code is open-source and available from https://github.com/huoxiangzuo/HiFuse.

Tianyuan Yao, Chang Qu, Jun Long et al.

With the rapid development of self-supervised learning (e.g., contrastive learning), the importance of having large-scale images (even without annotations) for training a more generalizable AI model has been widely recognized in medical image analysis. However, collecting large-scale task-specific unannotated data at scale can be challenging for individual labs. Existing online resources, such as digital books, publications, and search engines, provide a new resource for obtaining large-scale images. However, published images in healthcare (e.g., radiology and pathology) consist of a considerable amount of compound figures with subplots. In order to extract and separate compound figures into usable individual images for downstream learning, we propose a simple compound figure separation (SimCFS) framework without using the traditionally required detection bounding box annotations, with a new loss function and a hard case simulation. Our technical contribution is four-fold: (1) we introduce a simulation-based training framework that minimizes the need for resource extensive bounding box annotations; (2) we propose a new side loss that is optimized for compound figure separation; (3) we propose an intra-class image augmentation method to simulate hard cases; and (4) to the best of our knowledge, this is the first study that evaluates the efficacy of leveraging self-supervised learning with compound image separation. From the results, the proposed SimCFS achieved state-of-the-art performance on the ImageCLEF 2016 Compound Figure Separation Database. The pretrained self-supervised learning model using large-scale mined figures improved the accuracy of downstream image classification tasks with a contrastive learning algorithm. The source code of SimCFS is made publicly available at https://github.com/hrlblab/ImageSeperation.

Tianyuan Yao, Yuzhe Lu, Jun Long et al.

The quantitative detection, segmentation, and characterization of glomeruli from high-resolution whole slide imaging (WSI) play essential roles in the computer-assisted diagnosis and scientific research in digital renal pathology. Historically, such comprehensive quantification requires extensive programming skills in order to be able to handle heterogeneous and customized computational tools. To bridge the gap of performing glomerular quantification for non-technical users, we develop the Glo-In-One toolkit to achieve holistic glomerular detection, segmentation, and characterization via a single line of command. Additionally, we release a large-scale collection of 30,000 unlabeled glomerular images to further facilitate the algorithmic development of self-supervised deep learning. The inputs of the Glo-In-One toolkit are WSIs, while the outputs are (1) WSI-level multi-class circle glomerular detection results (which can be directly manipulated with ImageScope), (2) glomerular image patches with segmentation masks, and (3) different lesion types. To leverage the performance of the Glo-In-One toolkit, we introduce self-supervised deep learning to glomerular quantification via large-scale web image mining. The GGS fine-grained classification model achieved a decent performance compared with baseline supervised methods while only using 10% of the annotated data. The glomerular detection achieved an average precision of 0.627 with circle representations, while the glomerular segmentation achieved a 0.955 patch-wise Dice Similarity Coefficient (DSC).

Mingxu Chai, Ziyu Shen, Chenyu Liu et al.

Block Diffusion Models (BDMs) support parallel generation, flexible-length output, and KV caching, making them promising for efficient document parsing. However, existing BDMs bind denoising and cache commitment to fixed block boundaries: parallelism shrinks during intra-block denoising, while generated tokens cannot be cached until the whole block is completed. Moreover, intra-block bidirectional denoising conflicts with inter-block autoregression, creating inconsistent information flow that can challenge structure-sensitive recognition. We propose the Prefix-Adaptive Block Diffusion Model (PA-BDM), which replaces intra-block bidirectional denoising with causal denoising from prefix to suffix and treats the block size as a maximum candidate range rather than a fixed commitment unit. PA-BDM uses Confidence-gated Structural Loss (CSL) to build low-entropy prefixes before extending training to longer continuations. During inference, Progressive Prefix Commitment (PPC) then dynamically commits the longest reliable prefix into the KV cache and resets the next candidate range from the updated prefix, restoring a large parallel decoding space at each step. Experiments show that the 3B PA-BDM achieves higher recognition scores on several benchmarks and improves inference throughput by 71.6\% over the 2.5B MinerU-Diffusion.

Dong Guo, Faming Wu, Feida Zhu et al. · pku

We present Seed1.5-VL, a vision-language foundation model designed to advance general-purpose multimodal understanding and reasoning. Seed1.5-VL is composed with a 532M-parameter vision encoder and a Mixture-of-Experts (MoE) LLM of 20B active parameters. Despite its relatively compact architecture, it delivers strong performance across a wide spectrum of public VLM benchmarks and internal evaluation suites, achieving the state-of-the-art performance on 38 out of 60 public benchmarks. Moreover, in agent-centric tasks such as GUI control and gameplay, Seed1.5-VL outperforms leading multimodal systems, including OpenAI CUA and Claude 3.7. Beyond visual and video understanding, it also demonstrates strong reasoning abilities, making it particularly effective for multimodal reasoning challenges such as visual puzzles. We believe these capabilities will empower broader applications across diverse tasks. In this report, we mainly provide a comprehensive review of our experiences in building Seed1.5-VL across model design, data construction, and training at various stages, hoping that this report can inspire further research. Seed1.5-VL is now accessible at https://www.volcengine.com/ (Volcano Engine Model ID: doubao-1-5-thinking-vision-pro-250428)

Ke Ma, Jun Long, Hongxiao Fei et al.

Pre-trained Vision-Language Models (VLMs) struggle with Zero-Shot Anomaly Detection (ZSAD) due to a critical adaptation gap: they lack the local inductive biases required for dense prediction and employ inflexible feature fusion paradigms. We address these limitations through an Architectural Co-Design framework that jointly refines feature representation and cross-modal fusion. Our method proposes a parameter-efficient Convolutional Low-Rank Adaptation (Conv-LoRA) adapter to inject local inductive biases for fine-grained representation, and introduces a Dynamic Fusion Gateway (DFG) that leverages visual context to adaptively modulate text prompts, enabling a powerful bidirectional fusion. Extensive experiments on diverse industrial and medical benchmarks demonstrate superior accuracy and robustness, validating that this synergistic co-design is critical for robustly adapting foundation models to dense perception tasks. The source code is available at https://github.com/cockmake/ACD-CLIP.

Fan Wang, Lei Luo, En Zhu et al.

Recent Multiple Object Tracking (MOT) methods have gradually attempted to integrate object detection and instance re-identification (Re-ID) into a united network to form a one-stage solution. Typically, these methods use two separated branches within a single network to accomplish detection and Re-ID respectively without studying the inter-relationship between them, which inevitably impedes the tracking performance. In this paper, we propose an online multi-object tracking framework based on a hierarchical single-branch network to solve this problem. Specifically, the proposed single-branch network utilizes an improved Hierarchical Online In-stance Matching (iHOIM) loss to explicitly model the inter-relationship between object detection and Re-ID. Our novel iHOIM loss function unifies the objectives of the two sub-tasks and encourages better detection performance and feature learning even in extremely crowded scenes. Moreover, we propose to introduce the object positions, predicted by a motion model, as region proposals for subsequent object detection, where the intuition is that detection results and motion predictions can complement each other in different scenarios. Experimental results on MOT16 and MOT20 datasets show that we can achieve state-of-the-art tracking performance, and the ablation study verifies the effectiveness of each proposed component.

Zhan Yang, Osolo Ian Raymond, WuQing Sun et al.

Due to its fast retrieval and storage efficiency capabilities, hashing has been widely used in nearest neighbor retrieval tasks. By using deep learning based techniques, hashing can outperform non-learning based hashing technique in many applications. However, we argue that the current deep learning based hashing methods ignore some critical problems (e.g., the learned hash codes are not discriminative due to the hashing methods being unable to discover rich semantic information and the training strategy having difficulty optimizing the discrete binary codes). In this paper, we propose a novel image hashing method, termed as \textbf{\underline{A}}symmetric \textbf{\underline{D}}eep \textbf{\underline{S}}emantic \textbf{\underline{Q}}uantization (\textbf{ADSQ}). \textbf{ADSQ} is implemented using three stream frameworks, which consist of one \emph{LabelNet} and two \emph{ImgNets}. The \emph{LabelNet} leverages the power of three fully-connected layers, which are used to capture rich semantic information between image pairs. For the two \emph{ImgNets}, they each adopt the same convolutional neural network structure, but with different weights (i.e., asymmetric convolutional neural networks). The two \emph{ImgNets} are used to generate discriminative compact hash codes. Specifically, the function of the \emph{LabelNet} is to capture rich semantic information that is used to guide the two \emph{ImgNets} in minimizing the gap between the real-continuous features and the discrete binary codes. Furthermore, \textbf{ADSQ} can utilize the most critical semantic information to guide the feature learning process and consider the consistency of the common semantic space and Hamming space. Experimental results on three benchmarks (i.e., CIFAR-10, NUS-WIDE, and ImageNet) demonstrate that the proposed \textbf{ADSQ} can outperforms current state-of-the-art methods.

Jun Long, WuQing Sun, Zhan Yang et al.

Human Activity Recognition (HAR) using deep neural network has become a hot topic in human-computer interaction. Machine can effectively identify human naturalistic activities by learning from a large collection of sensor data. Activity recognition is not only an interesting research problem, but also has many real-world practical applications. Based on the success of residual networks in achieving a high level of aesthetic representation of the automatic learning, we propose a novel \textbf{A}symmetric \textbf{R}esidual \textbf{N}etwork, named ARN. ARN is implemented using two identical path frameworks consisting of (1) a short time window, which is used to capture spatial features, and (2) a long time window, which is used to capture fine temporal features. The long time window path can be made very lightweight by reducing its channel capacity, yet still being able to learn useful temporal representations for activity recognition. In this paper, we mainly focus on proposing a new model to improve the accuracy of HAR. In order to demonstrate the effectiveness of ARN model, we carried out extensive experiments on benchmark datasets (i.e., OPPORTUNITY, UniMiB-SHAR) and compared with some conventional and state-of-the-art learning-based methods. Then, we discuss the influence of networks parameters on performance to provide insights about its optimization. Results from our experiments show that ARN is effective in recognizing human activities via wearable datasets.

Zhan Yang, Osolo Ian Raymond, Wuqing Sun et al.

With the rapid growth of multimedia data (e.g., image, audio and video etc.) on the web, learning-based hashing techniques such as Deep Supervised Hashing (DSH) have proven to be very efficient for large-scale multimedia search. The recent successes seen in Learning-based hashing methods are largely due to the success of deep learning-based hashing methods. However, there are some limitations to previous learning-based hashing methods (e.g., the learned hash codes containing repetitive and highly correlated information). In this paper, we propose a novel learning-based hashing method, named Deep Attention-guided Hashing (DAgH). DAgH is implemented using two stream frameworks. The core idea is to use guided hash codes which are generated by the hashing network of the first stream framework (called first hashing network) to guide the training of the hashing network of the second stream framework (called second hashing network). Specifically, in the first network, it leverages an attention network and hashing network to generate the attention-guided hash codes from the original images. The loss function we propose contains two components: the semantic loss and the attention loss. The attention loss is used to punish the attention network to obtain the salient region from pairs of images; in the second network, these attention-guided hash codes are used to guide the training of the second hashing network (i.e., these codes are treated as supervised labels to train the second network). By doing this, DAgH can make full use of the most critical information contained in images to guide the second hashing network in order to learn efficient hash codes in a true end-to-end fashion. Results from our experiments demonstrate that DAgH can generate high quality hash codes and it outperforms current state-of-the-art methods on three benchmark datasets, CIFAR-10, NUS-WIDE, and ImageNet.

Lei Zhu, Jun Long, Chengyuan Zhang et al.

Due to the rapid development of mobile Internet techniques, cloud computation and popularity of online social networking and location-based services, massive amount of multimedia data with geographical information is generated and uploaded to the Internet. In this paper, we propose a novel type of cross-modal multimedia retrieval called geo-multimedia cross-modal retrieval which aims to search out a set of geo-multimedia objects based on geographical distance proximity and semantic similarity between different modalities. Previous studies for cross-modal retrieval and spatial keyword search cannot address this problem effectively because they do not consider multimedia data with geo-tags and do not focus on this type of query. In order to address this problem efficiently, we present the definition of $k$NN geo-multimedia cross-modal query at the first time and introduce relevant conceptions such as cross-modal semantic representation space. To bridge the semantic gap between different modalities, we propose a method named cross-modal semantic matching which contains two important component, i.e., CorrProj and LogsTran, which aims to construct a common semantic representation space for cross-modal semantic similarity measurement. Besides, we designed a framework based on deep learning techniques to implement common semantic representation space construction. In addition, a novel hybrid indexing structure named GMR-Tree combining geo-multimedia data and R-Tree is presented and a efficient $k$NN search algorithm called $k$GMCMS is designed. Comprehensive experimental evaluation on real and synthetic dataset clearly demonstrates that our solution outperforms the-state-of-the-art methods.

Zhan Yang, Osolo Ian Raymond, ChengYuan Zhang et al.

Deep Convolutional Neural Networks (DCNNs) are currently popular in human activity recognition applications. However, in the face of modern artificial intelligence sensor-based games, many research achievements cannot be practically applied on portable devices. DCNNs are typically resource-intensive and too large to be deployed on portable devices, thus this limits the practical application of complex activity detection. In addition, since portable devices do not possess high-performance Graphic Processing Units (GPUs), there is hardly any improvement in Action Game (ACT) experience. Besides, in order to deal with multi-sensor collaboration, all previous human activity recognition models typically treated the representations from different sensor signal sources equally. However, distinct types of activities should adopt different fusion strategies. In this paper, a novel scheme is proposed. This scheme is used to train 2-bit Convolutional Neural Networks with weights and activations constrained to {-0.5,0,0.5}. It takes into account the correlation between different sensor signal sources and the activity types. This model, which we refer to as DFTerNet, aims at producing a more reliable inference and better trade-offs for practical applications. Our basic idea is to exploit quantization of weights and activations directly in pre-trained filter banks and adopt dynamic fusion strategies for different activity types. Experiments demonstrate that by using dynamic fusion strategy can exceed the baseline model performance by up to ~5% on activity recognition like OPPORTUNITY and PAMAP2 datasets. Using the quantization method proposed, we were able to achieve performances closer to that of full-precision counterpart. These results were also verified using the UniMiB-SHAR dataset. In addition, the proposed method can achieve ~9x acceleration on CPUs and ~11x memory saving.

Kamal Al-Sabahi, Zuping Zhang, Jun Long et al.

The fast-growing amount of information on the Internet makes the research in automatic document summarization very urgent. It is an effective solution for information overload. Many approaches have been proposed based on different strategies, such as latent semantic analysis (LSA). However, LSA, when applied to document summarization, has some limitations which diminish its performance. In this work, we try to overcome these limitations by applying statistic and linear algebraic approaches combined with syntactic and semantic processing of text. First, the part of speech tagger is utilized to reduce the dimension of LSA. Then, the weight of the term in four adjacent sentences is added to the weighting schemes while calculating the input matrix to take into account the word order and the syntactic relations. In addition, a new LSA-based sentence selection algorithm is proposed, in which the term description is combined with sentence description for each topic which in turn makes the generated summary more informative and diverse. To ensure the effectiveness of the proposed LSA-based sentence selection algorithm, extensive experiment on Arabic and English are done. Four datasets are used to evaluate the new model, Linguistic Data Consortium (LDC) Arabic Newswire-a corpus, Essex Arabic Summaries Corpus (EASC), DUC2002, and Multilingual MSS 2015 dataset. Experimental results on the four datasets show the effectiveness of the proposed model on Arabic and English datasets. It performs comprehensively better compared to the state-of-the-art methods.

Jun Long, Qunfeng Liu, Xinpan Yuan et al.

Image similarity measures play an important role in nearest neighbor search and duplicate detection for large-scale image datasets. Recently, Minwise Hashing (or Minhash) and its related hashing algorithms have achieved great performances in large-scale image retrieval systems. However, there are a large number of comparisons for image pairs in these applications, which may spend a lot of computation time and affect the performance. In order to quickly obtain the pairwise images that theirs similarities are higher than the specific threshold T (e.g., 0.5), we propose a dynamic threshold filter of Minwise Hashing for image similarity measures. It greatly reduces the calculation time by terminating the unnecessary comparisons in advance. We also find that the filter can be extended to other hashing algorithms, on when the estimator satisfies the binomial distribution, such as b-Bit Minwise Hashing, One Permutation Hashing, etc. In this pager, we use the Bag-of-Visual-Words (BoVW) model based on the Scale Invariant Feature Transform (SIFT) to represent the image features. We have proved that the filter is correct and effective through the experiment on real image datasets.

Jun Long, Lei Zhu, Zhan Yang et al.

Vast amount of multimedia data contains massive and multifarious social information which is used to construct large-scale social networks. In a complex social network, a character should be ideally denoted by one and only one vertex. However, it is pervasive that a character is denoted by two or more vertices with different names, thus it is usually considered as multiple, different characters. This problem causes incorrectness of results in network analysis and mining. The factual challenge is that character uniqueness is hard to correctly confirm due to lots of complicated factors, e.g. name changing and anonymization, leading to character duplication. Early, limited research has shown that previous methods depended overly upon supplementary attribute information from databases. In this paper, we propose a novel method to merge the character vertices which refer to as the same entity but are denoted with different names. With this method, we firstly build the relationship network among characters based on records of social activities participated, which are extracted from multimedia sources. Then define temporal activity paths (TAPs) for each character over time. After that, we measure similarity of the TAPs for any two characters. If the similarity is high enough, the two vertices should be considered to the same character. Based on TAPs, we can determine whether to merge the two character vertices. Our experiments shown that this solution can accurately confirm character uniqueness in large-scale social network.

Jun Long, Lei Zhu, Chengyuan Zhang et al.

Due to the advances in mobile computing and multimedia techniques, there are vast amount of multimedia data with geographical information collected in multifarious applications. In this paper, we propose a novel type of image search named interactive geo-tagged image search which aims to find out a set of images based on geographical proximity and similarity of visual content, as well as the preference of users. Existing approaches for spatial keyword query and geo-image query cannot address this problem effectively since they do not consider these three type of information together for query. In order to solve this challenge efficiently, we propose the definition of interactive top-$k$ geo-tagged image query and then present a framework including candidate search stage , interaction stage and termination stage. To enhance the searching efficiency in a large-scale database, we propose the candidate search algorithm named GI-SUPER Search based on a new notion called superior relationship and GIR-Tree, a novel index structure. Furthermore, two candidate selection methods are proposed for learning the preferences of the user during the interaction. At last, the termination procedure and estimation procedure are introduced in brief. Experimental evaluation on real multimedia dataset demonstrates that our solution has a really high performance.

Chengyuan Zhang, Yunwu Lin, Lei Zhu et al.

With advances in multimedia technologies and the proliferation of smart phone, digital cameras, storage devices, there are a rapidly growing massive amount of multimedia data collected in many applications such as multimedia retrieval and management system, in which the data element is composed of text, image, video and audio. Consequently, the study of multimedia near duplicate detection has attracted significant concern from research organizations and commercial communities. Traditional solution minwish hashing (\minwise) faces two challenges: expensive preprocessing time and lower comparison speed. Thus, this work first introduce a hashing method called one permutation hashing (\oph) to shun the costly preprocessing time. Based on \oph, a more efficient strategy group based one permutation hashing (\goph) is developed to deal with the high comparison time. Based on the fact that the similarity of most multimedia data is not very high, this work design an new hashing method namely hierarchical one permutation hashing (\hoph) to further improve the performance. Comprehensive experiments on real multimedia datasets clearly show that with similar accuracy \hoph is five to seven times faster than