Pengcheng Wu
Research Subjects
Research Interests
Pengcheng Wu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Gai Luo
Chengdu Research Base of Giant Panda Breeding, Sichuan Key Laboratory of Conservation Biology for Endangered Wildlife, Chengdu 610086, China
1 shared publication - Han Su
Sichuan Normal University
1 shared publication - Mengnan He
Chengdu Research Base of Giant Panda Breeding, The Conservation of Endangered Wildlife Key Laboratory of Sichuan Province, Chengdu, China
1 shared publication - Peng Chen
Chengdu Research Base of Giant Panda Breeding, The Conservation of Endangered Wildlife Key Laboratory of Sichuan Province, Chengdu, China
1 shared publication - Peng Min
School of Computer Science, Sichuan Normal University, Chengdu 610021, China
1 shared publication - Qingyue Min
Chengdu Research Base of Giant Panda Breeding, Sichuan Key Laboratory of Conservation Biology for Endangered Wildlife, Chengdu 610086, China
1 shared publication
Latest Publications
- Open Access
Swin-Panda: Behavior Recognition for Giant Pandas Based on Local Fine-Grained and Spatiotemporal Displacement Features
The giant panda, a rare and iconic species endemic to China, has attracted significant attention from both domestic and international researchers due to its crucial ecological role, unique cultural value, and distinct evolutionary history. While substantial progress has been made in the field of individual identification, behavior recognition remains underdeveloped, facing challenges such as the lack of dynamic temporal features and insufficient extraction of behavioral characteristics. To address these challenges, we propose the Swin-Panda model, which leverages transfer learning based on the Video Swin Transformer architecture within the mmaction2 framework. In addition, we introduce two novel modules: the Comprehensive Perception Auxiliary Module and the Spatiotemporal Shift Attention Module. These modules facilitate the extraction of local and spatiotemporal information, allowing the model to more effectively capture the behavioral and movement patterns of giant pandas. Experimental results on the PACV-8 dataset demonstrate that our model achieves an accuracy of 88.02%, outperforming several benchmark models. This approach significantly enhances behavior recognition accuracy, thereby contributing to the advancement of panda welfare and species conservation efforts.
Open AccessSwin-Panda: Behavior Recognition for Giant Pandas Based on Local Fine-Grained and Spatiotemporal Displacement Features
The giant panda, a rare and iconic species endemic to China, has attracted significant attention from both domestic and international researchers due to its crucial ecological role, unique cultural value, and distinct evolutionary history. While substantial progress has been made in the field of individual identification, behavior recognition remains underdeveloped, facing challenges such as the lack of dynamic temporal features and insufficient extraction of behavioral characteristics. To address these challenges, we propose the Swin-Panda model, which leverages transfer learning based on the Video Swin Transformer architecture within the mmaction2 framework. In addition, we introduce two novel modules: the Comprehensive Perception Auxiliary Module and the Spatiotemporal Shift Attention Module. These modules facilitate the extraction of local and spatiotemporal information, allowing the model to more effectively capture the behavioral and movement patterns of giant pandas. Experimental results on the PACV-8 dataset demonstrate that our model achieves an accuracy of 88.02%, outperforming several benchmark models. This approach significantly enhances behavior recognition accuracy, thereby contributing to the advancement of panda welfare and species conservation efforts.