Yue Feng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yue Feng has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Jie Qin
Nanjing University of Aeronautics and Astronautics
2 shared publications - Limin Wang
State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China
1 shared publication - Qiang Chen
Tencent WeChat, Guangzhou, China
1 shared publication - Rong Quan
Nanjing University of Aeronautics and Astronautics, Nanjing, China
1 shared publication - Yizhen Jia
Nanjing University of Aeronautics and Astronautics, Nanjing, China
1 shared publication - Zhengye Zhang
University of Science and Technology of China, Hefei, China
1 shared publication
Latest Publications
- 26 May 2026
- Published by Springer Nature in Machine Intelligence Research
- Vol. 23 (5) , 1012-1027
Training-free Dense Video Captioning with Large-scale Pretrained Models
- 26 May 2026
- Published by Springer Nature in Machine Intelligence Research
- Vol. 23 (5) , 1012-1027
Training-free Dense Video Captioning with Large-scale Pretrained Models
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 135-143
RefineTAD: Learning Proposal-free Refinement for Temporal Action Detection
Temporal action detection (TAD) aims to localize the start and end frames of actions in untrimmed videos, which is a challenging task due to the similarity of adjacent frames and the ambiguity of action boundaries. Previous methods often generate coarse proposals first and then perform proposal-based refinement, which is coupled with prior action detectors and leads to proposal-oriented offsets. However, this paradigm increases the training difficulty of the TAD model and is heavily influenced by the quantity and quality of the proposals. To address the above issues, we decouple the refinement process from conventional TAD methods and propose a learnable, proposal-free refinement method for fine boundary localization, named RefineTAD. We first propose a multi-level refinement module to generate multi-scale boundary offsets, score offsets and boundary-aware probability at each time point based on the feature pyramid. Then, we propose an offset focusing strategy to progressively refine the predicted results of TAD models in a coarse-to-fine manner with our multi-scale offsets. We perform extensive experiments on three challenging datasets and demonstrate that our RefineTAD significantly improves the state-of-the-art TAD methods with minimal computational overhead.
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 135-143
Open AccessRefineTAD: Learning Proposal-free Refinement for Temporal Action Detection
Temporal action detection (TAD) aims to localize the start and end frames of actions in untrimmed videos, which is a challenging task due to the similarity of adjacent frames and the ambiguity of action boundaries. Previous methods often generate coarse proposals first and then perform proposal-based refinement, which is coupled with prior action detectors and leads to proposal-oriented offsets. However, this paradigm increases the training difficulty of the TAD model and is heavily influenced by the quantity and quality of the proposals. To address the above issues, we decouple the refinement process from conventional TAD methods and propose a learnable, proposal-free refinement method for fine boundary localization, named RefineTAD. We first propose a multi-level refinement module to generate multi-scale boundary offsets, score offsets and boundary-aware probability at each time point based on the feature pyramid. Then, we propose an offset focusing strategy to progressively refine the predicted results of TAD models in a coarse-to-fine manner with our multi-scale offsets. We perform extensive experiments on three challenging datasets and demonstrate that our RefineTAD significantly improves the state-of-the-art TAD methods with minimal computational overhead.