Pose Depth Estimation for Pose Guided Human Video Generation
发布时间：2025-04-30
点击次数：
- 发布时间：
- 2025-04-30
- 论文名称：
- Pose Depth Estimation for Pose Guided Human Video Generation
- 发表刊物：
- The 5th International Symposium on Artificial Intelligence and Robotics
- 摘要：
- In the research of human video generation based on motion prediction, the predicted pose sequences obtained by human motion prediction are used as the conditions of generator to guide the generation of human video under given pose sequences. In the phase of video generation, we analyze the defects of the existing methods and propose a new video generation framework to improve the problems caused by the lack of depth information in the process of guiding the image generation of 2D human pose. In the experiment, the proposed method of human video generation is tested and verified by using public datasets. The effect of the proposed method is better than that of the baseline methods, which solves the ambiguity caused by the lack of depth information in video generation and verifies the effectiveness of the proposed method.
- 合写作者：
- Y Wang, X Wang, P Jiang and F Wang
- 是否译文：
- 否
- 发表时间：
- 2020-08-10