Abstract | ||
---|---|---|
Although 3D hand pose estimation has made significant progress in recent years with the development of the deep neural network, most learning-based methods require a large amount of labeled data that is time-consuming to collect. In this paper, we propose a dual-branch self-boosting framework for self-supervised 3D hand pose estimation from depth images. First, we adopt a simple yet effective image-to-image translation technology to generate realistic depth images from synthetic data for network pre-training. Second, we propose a dual-branch network to perform 3D hand model estimation and pixel-wise pose estimation in a decoupled way. Through a part-aware model-fitting loss, the network can be updated according to the fine-grained differences between the hand model and the unlabeled real image. Through an inter-branch loss, the two complementary branches can boost each other continuously during self-supervised learning. Furthermore, we adopt a refinement stage to better utilize the prior structure information in the estimated hand model for a more accurate and robust estimation. Our method outperforms previous self-supervised methods by a large margin without using paired multi-view images and achieves comparable results to strongly supervised methods. Besides, by adopting our regenerated pose annotations, the performance of the skeleton-based gesture recognition is significantly improved. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/TIP.2022.3192708 | IEEE TRANSACTIONS ON IMAGE PROCESSING |
Keywords | DocType | Volume |
Three-dimensional displays, Solid modeling, Pose estimation, Data models, Shape, Gesture recognition, Training, 3D hand pose estimation, self-supervised training, 3D hand model, gesture recognition | Journal | 31 |
Issue | ISSN | Citations |
1 | 1057-7149 | 0 |
PageRank | References | Authors |
0.34 | 0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Pengfei Ren | 1 | 4 | 5.29 |
Haifeng Sun | 2 | 68 | 27.77 |
Jiachang hao | 3 | 1 | 1.70 |
Qi Qi | 4 | 210 | 56.01 |
J. Wang | 5 | 479 | 95.23 |
Jianxin Liao | 6 | 457 | 82.08 |