Weimin Tan

Weimin Tan, Siyuan Chen, Bo Yan

Diffusion models have demonstrated excellent performance in image generation. Although various few-shot semantic segmentation (FSS) models with different network structures have been proposed, performance improvement has reached a bottleneck. This paper presents the first work to leverage the diffusion model for FSS task, called DifFSS. DifFSS, a novel FSS paradigm, can further improve the performance of the state-of-the-art FSS models by a large margin without modifying their network structure. Specifically, we utilize the powerful generation ability of diffusion models to generate diverse auxiliary support images by using the semantic mask, scribble or soft HED boundary of the support image as control conditions. This generation process simulates the variety within the class of the query image, such as color, texture variation, lighting, $etc$. As a result, FSS models can refer to more diverse support images, yielding more robust representations, thereby achieving a consistent improvement in segmentation performance. Extensive experiments on three publicly available datasets based on existing advanced FSS models demonstrate the effectiveness of the diffusion model for FSS task. Furthermore, we explore in detail the impact of different input settings of the diffusion model on segmentation performance. Hopefully, this completely new paradigm will bring inspiration to the study of FSS task integrated with AI-generated content. Code is available at https://github.com/TrinitialChan/DifFSS

Manyu Li, Ruian He, Chenxi Ma et al.

Multimodal large language models (MLLMs) show remarkable potential for scientific reasoning, yet their performance in specialized domains such as microscopy remains limited by the scarcity of domain-specific training data and the difficulty of encoding fine-grained expert knowledge into model parameters. To bridge the gap, we introduce MicroWorld, a framework that constructs a multimodal attributed property graph (MAPG) from large-scale scientific image--caption corpora and leverages it to augment MLLM reasoning at inference time without any domain-specific fine-tuning. MicroWorld extracts biomedical entities and relations via scispaCy or LLM-based triplet mining, aligns images and entities in a shared embedding space using Qwen3-VL-Embedding, and assembles a knowledge graph comprising approximately 111K nodes and 346K typed edges spanning eight relation categories. At inference time, a graph-augmented retrieval pipeline matches query entities to the MAPG and injects structured knowledge context into the MLLM prompt. On the MicroVQA benchmark, MicroWorld improves the reasoning performance of Qwen3-VL-8B-Instruct by 37.5%, outperforming GPT-5 by 13.0% to achieve a new state-of-the-art. Furthermore, it yields a 6.0% performance gain on the MicroBench benchmark. Extensive experiments demonstrate the enhanced generalization capability introduced by MicroWorld. A qualitative case study further reveals both the mechanisms through which structured knowledge improves reasoning and the failure modes that point to promising future directions. Code and data are available at https://github.com/ieellee/MicroWorld.

Xuhao Jiang, Weimin Tan, Tian Tan et al.

Image-based single-modality compression learning approaches have demonstrated exceptionally powerful encoding and decoding capabilities in the past few years , but suffer from blur and severe semantics loss at extremely low bitrates. To address this issue, we propose a multimodal machine learning method for text-guided image compression, in which the semantic information of text is used as prior information to guide image compression for better compression performance. We fully study the role of text description in different components of the codec, and demonstrate its effectiveness. In addition, we adopt the image-text attention module and image-request complement module to better fuse image and text features, and propose an improved multimodal semantic-consistent loss to produce semantically complete reconstructions. Extensive experiments, including a user study, prove that our method can obtain visually pleasing results at extremely low bitrates, and achieves a comparable or even better performance than state-of-the-art methods, even though these methods are at 2x to 4x bitrates of ours.

Shili Zhou, Ruian He, Weimin Tan et al.

Optical Flow Estimation aims to find the 2D dense motion field between two frames. Due to the limitation of model structures and training datasets, existing methods often rely too much on local clues and ignore the integrity of objects, resulting in fragmented motion estimation. Through theoretical analysis, we find the pre-trained large vision models are helpful in optical flow estimation, and we notice that the recently famous Segment Anything Model (SAM) demonstrates a strong ability to segment complete objects, which is suitable for solving the fragmentation problem. We thus propose a solution to embed the frozen SAM image encoder into FlowFormer to enhance object perception. To address the challenge of in-depth utilizing SAM in non-segmentation tasks like optical flow estimation, we propose an Optical Flow Task-Specific Adaption scheme, including a Context Fusion Module to fuse the SAM encoder with the optical flow context encoder, and a Context Adaption Module to adapt the SAM features for optical flow task with Learned Task-Specific Embedding. Our proposed SAMFlow model reaches 0.86/2.10 clean/final EPE and 3.55/12.32 EPE/F1-all on Sintel and KITTI-15 training set, surpassing Flowformer by 8.5%/9.9% and 13.2%/16.3%. Furthermore, our model achieves state-of-the-art performance on the Sintel and KITTI-15 benchmarks, ranking #1 among all two-frame methods on Sintel clean pass.

Manyu Li, Ruian He, Chenxi Ma et al.

Multimodal Large Language Models are increasingly applied to biomedical imaging, yet scientific reasoning for microscopy remains limited by the scarcity of large-scale, high-quality training data. We introduce MicroVQA++, a three-stage, large-scale and high-quality microscopy VQA corpus derived from the BIOMEDICA archive. Stage one bootstraps supervision from expert-validated figure-caption pairs sourced from peer-reviewed articles. Stage two applies HiCQA-Graph, a novel heterogeneous graph over images, captions, and QAs that fuses NLI-based textual entailment, CLIP-based vision-language alignment, and agent signals to identify and filter inconsistent samples. Stage three uses a MultiModal Large Language Model (MLLM) agent to generate multiple-choice questions (MCQ) followed by human screening. The resulting release comprises a large training split and a human-checked test split whose Bloom's level hard-sample distribution exceeds the MicroVQA benchmark. Our work delivers (i) a quality-controlled dataset that couples expert literature with graph-based filtering and human refinement; (ii) HiCQA-Graph, the first graph that jointly models (image, caption, QA) for cross-modal consistency filtering; (iii) evidence that careful data construction enables 4B-scale MLLMs to reach competitive microscopy reasoning performance (e.g., GPT-5) and achieve state-of-the-art performance among open-source MLLMs. Code and dataset will be released after the review process concludes.

Chenxi Ma, Bo Yan, Weimin Tan et al.

Recent studies of deep learning based stereo image super-resolution (StereoSR) have promoted the development of StereoSR. However, existing StereoSR models mainly concentrate on improving quantitative evaluation metrics and neglect the visual quality of super-resolved stereo images. To improve the perceptual performance, this paper proposes the first perception-oriented stereo image super-resolution approach by exploiting the feedback, provided by the evaluation on the perceptual quality of StereoSR results. To provide accurate guidance for the StereoSR model, we develop the first special stereo image super-resolution quality assessment (StereoSRQA) model, and further construct a StereoSRQA database. Extensive experiments demonstrate that our StereoSR approach significantly improves the perceptual quality and enhances the reliability of stereo images for disparity estimation.

Xuhao Jiang, Weimin Tan, Ri Cheng et al.

Under stereo settings, the performance of image JPEG artifacts removal can be further improved by exploiting the additional information provided by a second view. However, incorporating this information for stereo image JPEG artifacts removal is a huge challenge, since the existing compression artifacts make pixel-level view alignment difficult. In this paper, we propose a novel parallax transformer network (PTNet) to integrate the information from stereo image pairs for stereo image JPEG artifacts removal. Specifically, a well-designed symmetric bi-directional parallax transformer module is proposed to match features with similar textures between different views instead of pixel-level view alignment. Due to the issues of occlusions and boundaries, a confidence-based cross-view fusion module is proposed to achieve better feature fusion for both views, where the cross-view features are weighted with confidence maps. Especially, we adopt a coarse-to-fine design for the cross-view interaction, leading to better performance. Comprehensive experimental results demonstrate that our PTNet can effectively remove compression artifacts and achieves superior performance than other testing state-of-the-art methods.

Chenxi Ma, Bo Yan, Qing Lin et al.

Deep neural networks have greatly promoted the performance of single image super-resolution (SISR). Conventional methods still resort to restoring the single high-resolution (HR) solution only based on the input of image modality. However, the image-level information is insufficient to predict adequate details and photo-realistic visual quality facing large upscaling factors (x8, x16). In this paper, we propose a new perspective that regards the SISR as a semantic image detail enhancement problem to generate semantically reasonable HR image that are faithful to the ground truth. To enhance the semantic accuracy and the visual quality of the reconstructed image, we explore the multi-modal fusion learning in SISR by proposing a Text-Guided Super-Resolution (TGSR) framework, which can effectively utilize the information from the text and image modalities. Different from existing methods, the proposed TGSR could generate HR image details that match the text descriptions through a coarse-to-fine process. Extensive experiments and ablation studies demonstrate the effect of the TGSR, which exploits the text reference to recover realistic images.

Shili Zhou, Xuhao Jiang, Weimin Tan et al.

In recent years, many deep learning-based methods have been proposed to tackle the problem of optical flow estimation and achieved promising results. However, they hardly consider that most videos are compressed and thus ignore the pre-computed information in compressed video streams. Motion vectors, one of the compression information, record the motion of the video frames. They can be directly extracted from the compression code stream without computational cost and serve as a solid prior for optical flow estimation. Therefore, we propose an optical flow model, MVFlow, which uses motion vectors to improve the speed and accuracy of optical flow estimation for compressed videos. In detail, MVFlow includes a key Motion-Vector Converting Module, which ensures that the motion vectors can be transformed into the same domain of optical flow and then be utilized fully by the flow estimation module. Meanwhile, we construct four optical flow datasets for compressed videos containing frames and motion vectors in pairs. The experimental results demonstrate the superiority of our proposed MVFlow, which can reduce the AEPE by 1.09 compared to existing models or save 52% time to achieve similar accuracy to existing models.

Ri Cheng, Yuqi Sun, Bo Yan et al.

Recent multi-view multimedia applications struggle between high-resolution (HR) visual experience and storage or bandwidth constraints. Therefore, this paper proposes a Multi-View Image Super-Resolution (MVISR) task. It aims to increase the resolution of multi-view images captured from the same scene. One solution is to apply image or video super-resolution (SR) methods to reconstruct HR results from the low-resolution (LR) input view. However, these methods cannot handle large-angle transformations between views and leverage information in all multi-view images. To address these problems, we propose the MVSRnet, which uses geometry information to extract sharp details from all LR multi-view to support the SR of the LR input view. Specifically, the proposed Geometry-Aware Reference Synthesis module in MVSRnet uses geometry information and all multi-view LR images to synthesize pixel-aligned HR reference images. Then, the proposed Dynamic High-Frequency Search network fully exploits the high-frequency textural details in reference images for SR. Extensive experiments on several benchmarks show that our method significantly improves over the state-of-the-art approaches.

Yuqi Sun, Ruian He, Weimin Tan et al.

Recent neural talking radiance field methods have shown great success in photorealistic audio-driven talking face synthesis. In this paper, we propose a novel interactive framework that utilizes human instructions to edit such implicit neural representations to achieve real-time personalized talking face generation. Given a short speech video, we first build an efficient talking radiance field, and then apply the latest conditional diffusion model for image editing based on the given instructions and guiding implicit representation optimization towards the editing target. To ensure audio-lip synchronization during the editing process, we propose an iterative dataset updating strategy and utilize a lip-edge loss to constrain changes in the lip region. We also introduce a lightweight refinement network for complementing image details and achieving controllable detail generation in the final rendered image. Our method also enables real-time rendering at up to 30FPS on consumer hardware. Multiple metrics and user verification show that our approach provides a significant improvement in rendering quality compared to state-of-the-art methods.

Ri Cheng, Xuhao Jiang, Ruian He et al.

The video frame interpolation (VFI) model applies the convolution operation to all locations, leading to redundant computations in regions with easy motion. We can use dynamic spatial pruning method to skip redundant computation, but this method cannot properly identify easy regions in VFI tasks without supervision. In this paper, we develop an Uncertainty-Guided Spatial Pruning (UGSP) architecture to skip redundant computation for efficient frame interpolation dynamically. Specifically, pixels with low uncertainty indicate easy regions, where the calculation can be reduced without bringing undesirable visual results. Therefore, we utilize uncertainty-generated mask labels to guide our UGSP in properly locating the easy region. Furthermore, we propose a self-contrast training strategy that leverages an auxiliary non-pruning branch to improve the performance of our UGSP. Extensive experiments show that UGSP maintains performance but reduces FLOPs by 34%/52%/30% compared to baseline without pruning on Vimeo90K/UCF101/MiddleBury datasets. In addition, our method achieves state-of-the-art performance with lower FLOPs on multiple benchmarks.

Ruian He, Zhen Xing, Weimin Tan et al.

Self-supervised representation learning has gained increasing attention for strong generalization ability without relying on paired datasets. However, it has not been explored sufficiently for facial representation. Self-supervised facial representation learning remains unsolved due to the coupling of facial identities, expressions, and external factors like pose and light. Prior methods primarily focus on contrastive learning and pixel-level consistency, leading to limited interpretability and suboptimal performance. In this paper, we propose LatentFace, a novel generative framework for self-supervised facial representations. We suggest that the disentangling problem can be also formulated as generative objectives in space and time, and propose the solution using a 3D-aware latent diffusion model. First, we introduce a 3D-aware autoencoder to encode face images into 3D latent embeddings. Second, we propose a novel representation diffusion model to disentangle 3D latent into facial identity and expression. Consequently, our method achieves state-of-the-art performance in facial expression recognition (FER) and face verification among self-supervised facial representation learning models. Our model achieves a 3.75\% advantage in FER accuracy on RAF-DB and 3.35\% on AffectNet compared to SOTA methods.

Jianzhi Lu, Ruian He, Shili Zhou et al.

Facial movements play a crucial role in conveying altitude and intentions, and facial optical flow provides a dynamic and detailed representation of it. However, the scarcity of datasets and a modern baseline hinders the progress in facial optical flow research. This paper proposes FacialFlowNet (FFN), a novel large-scale facial optical flow dataset, and the Decomposed Facial Flow Model (DecFlow), the first method capable of decomposing facial flow. FFN comprises 9,635 identities and 105,970 image pairs, offering unprecedented diversity for detailed facial and head motion analysis. DecFlow features a facial semantic-aware encoder and a decomposed flow decoder, excelling in accurately estimating and decomposing facial flow into head and expression components. Comprehensive experiments demonstrate that FFN significantly enhances the accuracy of facial flow estimation across various optical flow methods, achieving up to an 11% reduction in Endpoint Error (EPE) (from 3.91 to 3.48). Moreover, DecFlow, when coupled with FFN, outperforms existing methods in both synthetic and real-world scenarios, enhancing facial expression analysis. The decomposed expression flow achieves a substantial accuracy improvement of 18% (from 69.1% to 82.1%) in micro-expressions recognition. These contributions represent a significant advancement in facial motion analysis and optical flow estimation. Codes and datasets can be found.

Xuze Hao, Wenqian Ni, Xuhao Jiang et al.

Deep convolutional neural networks have made significant breakthroughs in medical image classification, under the assumption that training samples from all classes are simultaneously available. However, in real-world medical scenarios, there's a common need to continuously learn about new diseases, leading to the emerging field of class incremental learning (CIL) in the medical domain. Typically, CIL suffers from catastrophic forgetting when trained on new classes. This phenomenon is mainly caused by the imbalance between old and new classes, and it becomes even more challenging with imbalanced medical datasets. In this work, we introduce two simple yet effective plug-in methods to mitigate the adverse effects of the imbalance. First, we propose a CIL-balanced classification loss to mitigate the classifier bias toward majority classes via logit adjustment. Second, we propose a distribution margin loss that not only alleviates the inter-class overlap in embedding space but also enforces the intra-class compactness. We evaluate the effectiveness of our method with extensive experiments on three benchmark datasets (CCH5000, HAM10000, and EyePACS). The results demonstrate that our approach outperforms state-of-the-art methods.

Wenxuan Yang, Hanyu Zhang, Weimin Tan et al.

Self-supervised pre-training medical foundation models on large-scale datasets demonstrate exceptional performance. Recent research challenges this common paradigm by introducing data-effective learning approaches, demonstrating that merely increasing pre-training data volume does not necessarily improve model performance. However, current methods still have unclear standards and the underlying theoretical foundation remains unknown. In this paper, as the first attempt to address this limitation, we introduce V-information into self-supervised pre-training of foundation models to provide a theoretical foundation for sample selection. Our derivation confirms that by optimizing V-information, sample selection can be framed as an optimization problem where choosing diverse and challenging samples enhances model performance even under limited training data. Under this guidance, we develop an optimized data-effective learning method (OptiDEL) to optimize V-information in real-world medical domains by generating more diverse and harder samples. We compare the OptiDEL method with state-of-the-art approaches finding that OptiDEL consistently outperforms existing approaches across eight different datasets, with foundation models trained on only 5% of the pre-training data achieving up to 6.2% higher mIoU than those trained on the full dataset. Remarkably, OptiDEL demonstrates an average improvement of 4.7% mIoU over competing methods while using 20x less training data.

Wenqi Zeng, Yuqi Sun, Chenxi Ma et al.

Medical vision-language models (VLMs) have shown promise as clinical assistants across various medical fields. However, specialized dermatology VLM capable of delivering professional and detailed diagnostic analysis remains underdeveloped, primarily due to less specialized text descriptions in current dermatology multimodal datasets. To address this issue, we propose MM-Skin, the first large-scale multimodal dermatology dataset that encompasses 3 imaging modalities, including clinical, dermoscopic, and pathological and nearly 10k high-quality image-text pairs collected from professional textbooks. In addition, we generate over 27k diverse, instruction-following vision question answering (VQA) samples (9 times the size of current largest dermatology VQA dataset). Leveraging public datasets and MM-Skin, we developed SkinVL, a dermatology-specific VLM designed for precise and nuanced skin disease interpretation. Comprehensive benchmark evaluations of SkinVL on VQA, supervised fine-tuning (SFT) and zero-shot classification tasks across 8 datasets, reveal its exceptional performance for skin diseases in comparison to both general and medical VLM models. The introduction of MM-Skin and SkinVL offers a meaningful contribution to advancing the development of clinical dermatology VLM assistants. MM-Skin is available at https://github.com/ZwQ803/MM-Skin

Ruian He, Zhen Xing, Weimin Tan et al.

Affective Analysis is not a single task, and the valence-arousal value, expression class, and action unit can be predicted at the same time. Previous researches did not pay enough attention to the entanglement and hierarchical relation of these three facial attributes. We propose a novel model named feature pyramid networks for multi-task affect analysis. The hierarchical features are extracted to predict three labels and we apply a teacher-student training strategy to learn from pretrained single-task models. Extensive experiment results demonstrate the proposed model outperforms other models. This is a submission to The 2nd Workshop and Competition on Affective Behavior Analysis in the wild (ABAW). The code and model are available for research purposes at https://github.com/ryanhe312/ABAW2-FPNMAA.

Weimin Tan, Bo Yan, Chumin Lin et al.

Supervised deep learning techniques have achieved great success in various fields due to getting rid of the limitation of handcrafted representations. However, most previous image retargeting algorithms still employ fixed design principles such as using gradient map or handcrafted features to compute saliency map, which inevitably restricts its generality. Deep learning techniques may help to address this issue, but the challenging problem is that we need to build a large-scale image retargeting dataset for the training of deep retargeting models. However, building such a dataset requires huge human efforts. In this paper, we propose a novel deep cyclic image retargeting approach, called Cycle-IR, to firstly implement image retargeting with a single deep model, without relying on any explicit user annotations. Our idea is built on the reverse mapping from the retargeted images to the given input images. If the retargeted image has serious distortion or excessive loss of important visual information, the reverse mapping is unlikely to restore the input image well. We constrain this forward-reverse consistency by introducing a cyclic perception coherence loss. In addition, we propose a simple yet effective image retargeting network (IRNet) to implement the image retargeting process. Our IRNet contains a spatial and channel attention layer, which is able to discriminate visually important regions of input images effectively, especially in cluttered images. Given arbitrary sizes of input images and desired aspect ratios, our Cycle-IR can produce visually pleasing target images directly. Extensive experiments on the standard RetargetMe dataset show the superiority of our Cycle-IR. In addition, our Cycle-IR outperforms the Multiop method and obtains the best result in the user study. Code is available at https://github.com/mintanwei/Cycle-IR.

Wenxuan Yang, Weimin Tan, Yuqi Sun et al.

Foundation models, pre-trained on massive datasets, have achieved unprecedented generalizability. However, is it truly necessary to involve such vast amounts of data in pre-training, consuming extensive computational resources? This paper introduces data-effective learning, aiming to use data in the most impactful way to pre-train foundation models. This involves strategies that focus on data quality rather than quantity, ensuring the data used for training has high informational value. Data-effective learning plays a profound role in accelerating foundation model training, reducing computational costs, and saving data storage, which is very important as the volume of medical data in recent years has grown beyond many people's expectations. However, due to the lack of standards and comprehensive benchmarks, research on medical data-effective learning is poorly studied. To address this gap, our paper introduces a comprehensive benchmark specifically for evaluating data-effective learning in the medical field. This benchmark includes a dataset with millions of data samples from 31 medical centers (DataDEL), a baseline method for comparison (MedDEL), and a new evaluation metric (NormDEL) to objectively measure data-effective learning performance. Our extensive experimental results show the baseline MedDEL can achieve performance comparable to the original large dataset with only 5% of the data. Establishing such an open data-effective learning benchmark is crucial for the medical foundation model research community because it facilitates efficient data use, promotes collaborative breakthroughs, and fosters the development of cost-effective, scalable, and impactful healthcare solutions.

Ruian He, Shili Zhou, Yuqi Sun et al.

With the rise of real-time rendering and the evolution of display devices, there is a growing demand for post-processing methods that offer high-resolution content in a high frame rate. Existing techniques often suffer from quality and latency issues due to the disjointed treatment of frame supersampling and extrapolation. In this paper, we recognize the shared context and mechanisms between frame supersampling and extrapolation, and present a novel framework, Space-time Supersampling (STSS). By integrating them into a unified framework, STSS can improve the overall quality with lower latency. To implement an efficient architecture, we treat the aliasing and warping holes unified as reshading regions and put forth two key components to compensate the regions, namely Random Reshading Masking (RRM) and Efficient Reshading Module (ERM). Extensive experiments demonstrate that our approach achieves superior visual fidelity compared to state-of-the-art (SOTA) methods. Notably, the performance is achieved within only 4ms, saving up to 75\% of time against the conventional two-stage pipeline that necessitates 17ms.

Ri Cheng, Ruian He, Xuhao Jiang et al.

Existing recurrent optical flow estimation networks are computationally expensive since they use a fixed large number of iterations to update the flow field for each sample. An efficient network should skip iterations when the flow improvement is limited. In this paper, we develop a Context-Aware Iteration Policy Network for efficient optical flow estimation, which determines the optimal number of iterations per sample. The policy network achieves this by learning contextual information to realize whether flow improvement is bottlenecked or minimal. On the one hand, we use iteration embedding and historical hidden cell, which include previous iterations information, to convey how flow has changed from previous iterations. On the other hand, we use the incremental loss to make the policy network implicitly perceive the magnitude of optical flow improvement in the subsequent iteration. Furthermore, the computational complexity in our dynamic network is controllable, allowing us to satisfy various resource preferences with a single trained model. Our policy network can be easily integrated into state-of-the-art optical flow networks. Extensive experiments show that our method maintains performance while reducing FLOPs by about 40%/20% for the Sintel/KITTI datasets.

Yu Ling, Weimin Tan, Bo Yan

Survival analysis aims at modeling the relationship between covariates and event occurrence with some untracked (censored) samples. In implementation, existing methods model the survival distribution with strong assumptions or in a discrete time space for likelihood estimation with censorship, which leads to weak generalization. In this paper, we propose Implicit Survival Function (ISF) based on Implicit Neural Representation for survival distribution estimation without strong assumptions,and employ numerical integration to approximate the cumulative distribution function for prediction and optimization. Experimental results show that ISF outperforms the state-of-the-art methods in three public datasets and has robustness to the hyperparameter controlling estimation precision.

Xuhao Jiang, Weimin Tan, Qing Lin et al.

In recent years, many convolutional neural network-based models are designed for JPEG artifacts reduction, and have achieved notable progress. However, few methods are suitable for extreme low-bitrate image compression artifacts reduction. The main challenge is that the highly compressed image loses too much information, resulting in reconstructing high-quality image difficultly. To address this issue, we propose a multimodal fusion learning method for text-guided JPEG artifacts reduction, in which the corresponding text description not only provides the potential prior information of the highly compressed image, but also serves as supplementary information to assist in image deblocking. We fuse image features and text semantic features from the global and local perspectives respectively, and design a contrastive loss built upon contrastive learning to produce visually pleasing results. Extensive experiments, including a user study, prove that our method can obtain better deblocking results compared to the state-of-the-art methods.

Dawei Dai, Weimin Tan, Hong Zhan

In recent years, deep learning based on artificial neural network (ANN) has achieved great success in pattern recognition. However, there is no clear understanding of such neural computational models. In this paper, we try to unravel "black-box" structure of Ann model from network flow. Specifically, we consider the feed forward Ann as a network flow model, which consists of many directional class-pathways. Each class-pathway encodes one class. The class-pathway of a class is obtained by connecting the activated neural nodes in each layer from input to output, where activation value of neural node (node-value) is defined by the weights of each layer in a trained ANN-classifier. From the perspective of the class-pathway, training an ANN-classifier can be regarded as the formulation process of class-pathways of different classes. By analyzing the the distances of each two class-pathways in a trained ANN-classifiers, we try to answer the questions, why the classifier performs so? At last, from the neural encodes view, we define the importance of each neural node through the class-pathways, which is helpful to optimize the structure of a classifier. Experiments for two types of ANN model including multi-layer MLP and CNN verify that the network flow based on class-pathway is a reasonable explanation for ANN models.