Publications


Prototypical Cross-Attention Networks for Multiple Object Tracking and Segmentation.
In Proceedings of Advances in Neural Information Processing Systems (NeurIPS),  2021.

PDF Code Video


Towards efficient scene understanding via squeeze reasoning.
In Proceedings of IEEE Transactions on Image Processing (TIP),  2021.

PDF Code


Quasi-dense similarity learning for multiple object tracking.
In Proceedings of Conference on Computer Vision and Pattern Recognition (CVPR),  2021.

PDF Code


PointFlow: Flowing Semantics Through Points for Aerial Image Segmentation.
In Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR),  2021.

PDF Code


Bi-directional Exponential Angular Triplet Loss for RGB-Infrared Person Re-Identification.
In Proceedings of Transactions on Image Processsing (TIP),  2020.

PDF Code Zhihu


Is Attention Better Than Matrix Decomposition?.
In Proceedings of International Conference on Learning Representations (ICLR),  2020.

PDF Code Zhihu


Improving Semantic Segmentation via Decoupled Body and Edge Supervision.
In Proceedings of European Conference on Computer Vision (ECCV),  2020.

PDF Code


Spatial Pyramid Based Graph Reasoning for Semantic Segmentation.
In Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR),  2020.

PDF


SOGNet: Scene Overlap Graph Network for Panoptic Segmentation.
In Proceedings of AAAI Conference on Aritifical Intelligence (AAAI),  2020.

PDF Code Slides


Dynamic System Inspired Adaptive Time Stepping Controller for Residual Networks Families.
In Proceedings of AAAI Conference on Artificial Intelligence (AAAI),  2020.

PDF


Expectation Maximization Attention Networks for Semantic Segmentation.
In Proceedings of International Conference in Computer Vision (ICCV),  2019.  (Oral 4.3%).

PDF Code Project Slides Zhihu Survey Leaderboard


R^2 Net Recurrent and Recursive Network for Sparse View CT Artifacts Removal.
In Proceedings of Medical Image Computing and Computer Assisted Interventions (MICCAI),  2019.  (equal contribution).


Self Refining Deep Symmetry Enhanced Network for Rain Removal.
In Proceedings of International Conference on Image Processing (ICIP),  2019.

PDF Code


Recurrent Squeeze-and-Excitation Net for Single Image Deraining.
In Proceedings of European Conference on Computer Vision (ECCV),  2018.

PDF Code Dataset Report