SDPose: Tokenized Pose Estimation via Circulation-Guide Self-Distillation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916193297760256 |
|---|---|
| author | Chen, Sichen Zhang, Yingyi Huang, Siming Yi, Ran Fan, Ke Zhang, Ruixin Chen, Peixian Wang, Jun Ding, Shouhong Ma, Lizhuang |
| author_facet | Chen, Sichen Zhang, Yingyi Huang, Siming Yi, Ran Fan, Ke Zhang, Ruixin Chen, Peixian Wang, Jun Ding, Shouhong Ma, Lizhuang |
| contents | Recently, transformer-based methods have achieved state-of-the-art prediction quality on human pose estimation(HPE). Nonetheless, most of these top-performing transformer-based models are too computation-consuming and storage-demanding to deploy on edge computing platforms. Those transformer-based models that require fewer resources are prone to under-fitting due to their smaller scale and thus perform notably worse than their larger counterparts. Given this conundrum, we introduce SDPose, a new self-distillation method for improving the performance of small transformer-based models. To mitigate the problem of under-fitting, we design a transformer module named Multi-Cycled Transformer(MCT) based on multiple-cycled forwards to more fully exploit the potential of small model parameters. Further, in order to prevent the additional inference compute-consuming brought by MCT, we introduce a self-distillation scheme, extracting the knowledge from the MCT module to a naive forward model. Specifically, on the MSCOCO validation dataset, SDPose-T obtains 69.7% mAP with 4.4M parameters and 1.8 GFLOPs. Furthermore, SDPose-S-V2 obtains 73.5% mAP on the MSCOCO validation dataset with 6.2M parameters and 4.7 GFLOPs, achieving a new state-of-the-art among predominant tiny neural network methods. Our code is available at https://github.com/MartyrPenink/SDPose. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_03518 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SDPose: Tokenized Pose Estimation via Circulation-Guide Self-Distillation Chen, Sichen Zhang, Yingyi Huang, Siming Yi, Ran Fan, Ke Zhang, Ruixin Chen, Peixian Wang, Jun Ding, Shouhong Ma, Lizhuang Computer Vision and Pattern Recognition Recently, transformer-based methods have achieved state-of-the-art prediction quality on human pose estimation(HPE). Nonetheless, most of these top-performing transformer-based models are too computation-consuming and storage-demanding to deploy on edge computing platforms. Those transformer-based models that require fewer resources are prone to under-fitting due to their smaller scale and thus perform notably worse than their larger counterparts. Given this conundrum, we introduce SDPose, a new self-distillation method for improving the performance of small transformer-based models. To mitigate the problem of under-fitting, we design a transformer module named Multi-Cycled Transformer(MCT) based on multiple-cycled forwards to more fully exploit the potential of small model parameters. Further, in order to prevent the additional inference compute-consuming brought by MCT, we introduce a self-distillation scheme, extracting the knowledge from the MCT module to a naive forward model. Specifically, on the MSCOCO validation dataset, SDPose-T obtains 69.7% mAP with 4.4M parameters and 1.8 GFLOPs. Furthermore, SDPose-S-V2 obtains 73.5% mAP on the MSCOCO validation dataset with 6.2M parameters and 4.7 GFLOPs, achieving a new state-of-the-art among predominant tiny neural network methods. Our code is available at https://github.com/MartyrPenink/SDPose. |
| title | SDPose: Tokenized Pose Estimation via Circulation-Guide Self-Distillation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2404.03518 |