Jing Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jing Li has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Dongze Lian
National University of Singapore
1 shared publication - Huang
ShanghaiTech University
1 shared publication - Jinpeng Yu
School of Information Science and Technology, ShanghaiTech University, Shanghai, China
1 shared publication - Ruoyu Wang
School of Information Science and Technology, ShanghaiTech University, Shanghai, China
1 shared publication Shenghua Gao
ShanghaiTech University
1 shared publication- Shenghua Gao
The University of Hong Kong
1 shared publication
Latest Publications
- 31 December 2024
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 133 (6) , 3203-3221
Pseudo-Plane Regularized Signed Distance Field for Neural Indoor Scene Reconstruction
- 31 December 2024
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 133 (6) , 3203-3221
Pseudo-Plane Regularized Signed Distance Field for Neural Indoor Scene Reconstruction
- Open Access
TSP-Transformer: Task-Specific Prompts Boosted Transformer for Holistic Scene Understanding
Holistic scene understanding includes semantic segmentation, surface normal estimation, object boundary detection, depth estimation, etc. The key aspect of this problem is to learn representation effectively, as each subtask builds upon not only correlated but also distinct attributes. Inspired by visual-prompt tuning, we propose a Task-Specific Prompts Transformer, dubbed TSP-Transformer, for holistic scene understanding. It features a vanilla transformer in the early stage and tasks-specific prompts transformer encoder in the lateral stage, where tasks-specific prompts are augmented. By doing so, the transformer layer learns the generic information from the shared parts and is endowed with task-specific capacity. First, the tasks-specific prompts serve as induced priors for each task effectively. Moreover, the task-specific prompts can be seen as switches to favor task-specific representation learning for different tasks. Extensive experiments on NYUD-v2 and PASCAL-Context show that our method achieves state-of-the-art performance, validating the effectiveness of our method for holistic scene understanding. We also provide our code in the following link https://github.com/tb2-sy/TSP-Transformer.
Open AccessTSP-Transformer: Task-Specific Prompts Boosted Transformer for Holistic Scene Understanding
Holistic scene understanding includes semantic segmentation, surface normal estimation, object boundary detection, depth estimation, etc. The key aspect of this problem is to learn representation effectively, as each subtask builds upon not only correlated but also distinct attributes. Inspired by visual-prompt tuning, we propose a Task-Specific Prompts Transformer, dubbed TSP-Transformer, for holistic scene understanding. It features a vanilla transformer in the early stage and tasks-specific prompts transformer encoder in the lateral stage, where tasks-specific prompts are augmented. By doing so, the transformer layer learns the generic information from the shared parts and is endowed with task-specific capacity. First, the tasks-specific prompts serve as induced priors for each task effectively. Moreover, the task-specific prompts can be seen as switches to favor task-specific representation learning for different tasks. Extensive experiments on NYUD-v2 and PASCAL-Context show that our method achieves state-of-the-art performance, validating the effectiveness of our method for holistic scene understanding. We also provide our code in the following link https://github.com/tb2-sy/TSP-Transformer.