您当前的位置:
视觉建模组
组 长: 常虹 研究员
Email: changhong [at] ict dot ac dot cn; hong.chang [at] vipl dot ict dot ac dot cn
课题组简介

    视觉建模组致力于研究新的机器学习方法与技术,以解决现实的计算机视觉问题,主要包括图像视频的表示、理解、预测和增强,以及面向下一代智能视频监控的目标(特别是行人)检测、跟踪、识别和检索。过去的几年中,视觉建模组在相关领域会议和期刊上发表多篇论文。最近,该组成员获得ECCV’2018行人检测竞赛冠军,以及ICME’18铂金最佳论文奖。

研究内容

机器学习的模型和算法,特别是

1) 度量学习
2) 小样本学习, 元学习
3) 无监督/半监督/弱监督学习
4) 深度学习


目标检测、跟踪与识别

1) 行人再识别
2) 目标(特别是行人)检测
3) 单目标/多目标跟踪


图像/视频表示学习

1) 属性学习
2) 视觉关系检测
3) 组合语义学习

图像/视频生成与增强

1) 图像生成
2) 视频预测
3) 图像/视频超分辨率



部分论文

刊物论文

  • Wei Zheng, Luhong Liang,Hong Chang, Cher-Keng Heng, Shiguang Shan, Xilin Chen, “Boosted Translation-Tolerable Classifiers for Fast Object Detection. Image and Vision Computing,” Image and Vision Computing, vol.30, no.8, pp.480-491,Aug.2012.
  • Deming Zhai, Hong Chang, Shiguang Shan, Xilin Chen, Wen Gao, \"Multi-View Metric Learning with Global Consistency and Local Smoothness,\" ACM Transactions on Intelligent Systems and Technology, Volume 3 Issue 3, May 2012.
  • Bingpeng Ma, Xiujuan Chai, Tianjiang Wang, \"A Novel Feature Descriptor Based on Biologically Inspired Feature for Head Pose Estimation,\" Neurocomputing, vol. 115, pp. 1-10, 2013.
  • Bingpeng Ma, Yu Su, Frederic Jurie, \"Covariance Descriptor Based on Bio-Inspired Features for Person Re-Identification and Face Verification,\" Image and Vision Computing, vol. 32, no. 6-7, pp. 379-390, 2014.
  • Wei Zheng, Hong Chang, Luhong Liang, Haoyu Ren, Shiguang Shan, Xilin Chen, “Strip Features for Fast Object Detection,” IEEE Transactions on Systems, Man and Cybernetics, Part B: Cybernetics, vol.43, no.6, pp.1898-1912,Dec.2013.
  • Zhen Cui, Hong Chang, Shiguang Shan, Bingpeng Ma, \"Joint Sparse Representation for Video-Based face Recognition,\" Neurocomputing, 2014.
  • Hao Liu, Bingpeng Ma, Lei Qin, Junbiao Pang, Chunjie Zhang, Qingming Huang, \"Set-Label Modeling and Deep Metric Learning on Person Re-Identification,\" Neurocomputing, vol. 151, no. 3, pp. 1283-1292, 2015.
  • Kongming Liang, Hong Chang, Bingpeng Ma, Shiguang Shan, Xilin Chen, \"Unifying Visual Attribute Learning with Object Recognition in a Multiplicative Framework,\" IEEE Transactions on Pattern Analysis and Machine Intelligence(TPAMI), 41(7):1747 - 1760, 2019.
  • Kongming Liang, Hong Chang, Shiguang Shan, Xilin Chen, "Visual Concept Conjunction Learning with Recurrent Neural Networks," Neurocomputing, Vol. 395, PP. 229-236, June 2020.
  • Yucheng Chen, Bingpeng Ma and Hong Chang, "Part Alignment Network for Vehicle Re-identification," Neurocomputing, vol. 418, no. 12, pp. 114-125, December 2020.
  • Furong Xu, Bingpeng Ma, Hong Chang and Shiguang Shan. "Isosceles Constraints for Person Re-identification." IEEE Trans. on Image Processing, vol. 29, no. 11, pp. 8930-8943, November 2020.
  • Fengling Mao, Bingpeng Ma, Hong Chang, Shiguang Shan and Xilin Chen. Learning Efficient Text to Image Synthesis via Interstage Cross-sample Similarity Distillation. SCIENCE CHINA Information Sciences, 64(2): 120102:1-120102:12, 2021.
  • Ruibing Hou, Bingpeng Ma, Hong Chang, Xinqian Gu, Shiguang Shan and Xilin Chen. IAUnet: Global Context-Aware Feature Learning for Person Re-Identification. IEEE Trans. on Neural Networks and Learning Systems (TNNLS), 32(10):4460-4474, 2021.
  • Xiangzhou Zhang, Bingpeng Ma, Hong Chang, Shiguang Shan and Xilin Chen. Location Sensitive Network for Human Instance Segmentation. IEEE Transactions on Image Processing (TIP), 30:7649-7662, 2021.
  • Yucheng Chen, Rui Huang, Hong Chang, Chuanqi Tan, Tao Xue and Bingpeng Ma. Cross-Modal Knowledge Adaptation for Language-Based Person Search. IEEE Transactions on Image Processing (TIP), 30:4057-4069, 2021.
  • Furong Xu, Bingpeng Ma, Hong Chang and Shiguang Shan. PRDP: Person Re-identification with Dirty and Poor Data. IEEE Transactions on Cybernetics, 2021. (Accepted)
  • Xiaoyi Yin, Zhen Cui, Hong Chang, Bingpeng Ma and Shiguang Shan. Extending Generalized Unsupervised Manifold Alignment. SCIENCE CHINA Information Sciences, 2021. (Accepted)
  • Ruibing Hou, Bingpeng Ma, Hong Chang, Xinqian Gu, Shiguang Shan and Xilin Chen. Feature Completion for Occluded Person Re-Identification. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021. (Accepted)
  • Cheng Wang, Bingpeng Ma, Hong Chang, Shiguang Shan and Xilin Chen. Person Search by a Bi-directional Task-Consistent Learning Model. IEEE Transactions on Multimedia (TMM), 2021. (Accepted)
  • Shutao Bai, Bingpeng Ma, Hong Chang, Rui Huang, Shiguang Shan and Xilin Chen. SANet: Statistic Attention Network for Person Re-Identification. IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2021. (Accepted)

会议论文

  • Xiaqing Xu, Bingpeng Ma, Hong Chang, Xilin Chen, \"Siamese Recurrent Architecture for Visual Tracking,\" IEEE International Conference on Image Processing(ICIP2017), 2017.
  • Kongming Liang, Yuhong Guo, Hong Chang, Xilin Chen, \"Visual Relationship Detection with Deep Structural Ranking,\" AAAI Conference on Artificial Intelligence(AAAI2018), New Orleans, Louisiana, USA, 2018.
  • Fangyi Zhang, Bingpeng Ma, Hong Chang, Shiguang Shan and Xilin Chen, \"Relation-aware Multiple Attention Siamese Networks for Robust Visual Tracking,\" British Machine Vision Conference (BMVC), Cardiff, UK, September, 9-12, 2019.
  • Shen Li, Bingpeng Ma, Hong Chang, Shiguang Shan, Xilin Chen, \"Continuity-Discrimination Convolutional Neural Network for Visual Object Tracking,\" IEEE International Conference on Multimedia and Expo, San Diego, USA, July 23-27,2018.
  • Furong Xu, Bingpeng Ma, Hong Chang, Shiguang Shan, Xilin Chen, \"Style Transfer with Adversarial Learning for Cross-Dataset Person Re-identification,\" Asian Conference on Computer Vision 2018(ACCV2018), 2-6 Dec. 2018, Perth Western Australia.
  • Ruibing Hou, Bingpeng Ma, Hong Chang, Xinqian Gu, Shiguang Shan, Xilin Chen, “Interaction-and-Aggregation Network for Person Re-identification,” IEEE Conference on Computer Vision and Pattern Recognition, pp.9317-9326, Long Beach, California, USA, June 16-20, 2019.
  • Ruibing Hou, Hong Chang, Bingpeng Ma and Xilin Chen, \"Video Prediction with Bidirectional Constraint Network,\" 14th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2019), pp. 1-8, Lille, France, May 14-18, 2019.
  • Ruibing Hou, Bingpeng Ma, Hong Chang, Xinqian Gu, Shiguang Shan, Xilin Chen, “VRSTC: Occlusion-Free Video Person Re-Identification,” IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.7183-7192, Long Beach, California, USA, June 16-20, 2019.
  • Liqiang Bao, Bingpeng Ma, Hong Chang, Xilin Chen, \"Masked Graph Attention Network for Person Re-identification,\" International Workshop conjunction with IEEE Computer Vision and Pattern Recognition(CVPR), Long Beach, California, USA, June 2019.
  • Liqiang Bao, Bingpeng Ma, Hong Chang, Xilin Chen, \"Preserving Structural Relationships for Person Re-identification,\" International Workshop conjunction with IEEE International Conference on Multimedia and Expo(ICME), pp.120-125, July 2019.
  • Hongkai Zhang, Hong Chang, Bingpeng Ma, Shiguang Shan and Xilin Chen, \"Cascade RetinaNet: Maintaining Consistency for Single-Stage Object Detection,\" British Machine Vision Conference (BMVC), Cardiff, UK, September, 9-12, 2019.
  • Xiaoyi Yin, Xinqian Gu, Hong Chang, Bingpeng Ma and Xilin Chen, \"Attribute-aware Pedestrian Image Editing,\" International Conference on Image and Graphics (ICIG), August 2019.
  • Fangyi Zhang, Bingpeng Ma, Hong Chang, Shiguang Shan and Xilin Chen, \"Relation-aware Multiple Attention Siamese Networks for Robust Visual Tracking,\" British Machine Vision Conference (BMVC), Cardiff, UK, September, 9-12, 2019.
  • Ruibing Hou, Hong Chang, Bingpeng Ma, Shiguang Shan, Xilin Chen, \"Cross Attention Network for Few-shot Classification,\"The Thirty-third Annual Conference on Neural Information Processing Systems (NeurIPS), pp.4005-4016, Vancouver, Canada, Dec. 8-14, 2019.
  • Fengling Mao, Bingpeng Ma, Hong Chang, Shiguang Shan and Xilin Chen, \"MS-GAN: Text to Image Synthesis with Attention-Modulated Generators and Similarity-aware Discriminators,\" British Machine Vision Conference (BMVC), Cardiff, UK, September, 9-12, 2019.
  • Hongkai Zhang, Hong Chang, Bingpeng Ma, Naiyan Wang, Xilin Chen, "Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training," Proceedings of the 16th European Conference on Computer Vision (ECCV), LNCS 12360, Vol.15, pp.260-275, Glasgow, UK / Cyberspace, August 23-28, 2020.
  • Cheng Wang, Bingpeng Ma, Hong Chang, Shiguang Shan, Xilin Chen, \"TCTS: A Task-Consistent Two-stage Framework for Person Search,\" IEEE Conference on Computer Vision and Pattern Recognition(CVPR 2020), pp. 11952-11961, 2020.
  • RuiBing Hou, Hong Chang, Bingpeng Ma, Shiguang Shan, Xilin Chen, "Temporal Complementary Learning for Video Person Re-Identification," Proceedings of the 16th European Conference on Computer Vision (ECCV), LNCS 12370, Vol.25, pp.388-405, Glasgow, UK / Cyberspace, August 23-28, 2020.
  • Xinqian Gu, Hong Chang, Bingpeng Ma, Hongkai Zhang, Xilin Chen, "Appearance-Preserving 3D Convolution for Video-based Person Re-identification," Proceedings of the 16th European Conference on Computer Vision (ECCV), LNCS 12347, Vol.2, pp.228-243, Glasgow, UK / Cyberspace, August 23-28, 2020.
  • Ruibing Hou, Hong Chang, Bingpeng Ma, Rui Huang and Shiguang Shan. BiCnet-TKS: Learning Efficient Spatial-Temporal Representation for Video Person Re-Identification. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2014–2023, Virtual Event, Jun. 19-25, 2021.