$\text{Di}^2\text{Pose}$: Discrete Diffusion Model for Occluded 3D Human Pose Estimation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911888981360640 |
|---|---|
| author | Wang, Weiquan Xiao, Jun Wang, Chunping Liu, Wei Wang, Zhao Chen, Long |
| author_facet | Wang, Weiquan Xiao, Jun Wang, Chunping Liu, Wei Wang, Zhao Chen, Long |
| contents | Continuous diffusion models have demonstrated their effectiveness in addressing the inherent uncertainty and indeterminacy in monocular 3D human pose estimation (HPE). Despite their strengths, the need for large search spaces and the corresponding demand for substantial training data make these models prone to generating biomechanically unrealistic poses. This challenge is particularly noticeable in occlusion scenarios, where the complexity of inferring 3D structures from 2D images intensifies. In response to these limitations, we introduce the Discrete Diffusion Pose ($\text{Di}^2\text{Pose}$), a novel framework designed for occluded 3D HPE that capitalizes on the benefits of a discrete diffusion model. Specifically, $\text{Di}^2\text{Pose}$ employs a two-stage process: it first converts 3D poses into a discrete representation through a \emph{pose quantization step}, which is subsequently modeled in latent space through a \emph{discrete diffusion process}. This methodological innovation restrictively confines the search space towards physically viable configurations and enhances the model's capability to comprehend how occlusions affect human pose within the latent space. Extensive evaluations conducted on various benchmarks (e.g., Human3.6M, 3DPW, and 3DPW-Occ) have demonstrated its effectiveness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_17016 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | $\text{Di}^2\text{Pose}$: Discrete Diffusion Model for Occluded 3D Human Pose Estimation Wang, Weiquan Xiao, Jun Wang, Chunping Liu, Wei Wang, Zhao Chen, Long Computer Vision and Pattern Recognition Continuous diffusion models have demonstrated their effectiveness in addressing the inherent uncertainty and indeterminacy in monocular 3D human pose estimation (HPE). Despite their strengths, the need for large search spaces and the corresponding demand for substantial training data make these models prone to generating biomechanically unrealistic poses. This challenge is particularly noticeable in occlusion scenarios, where the complexity of inferring 3D structures from 2D images intensifies. In response to these limitations, we introduce the Discrete Diffusion Pose ($\text{Di}^2\text{Pose}$), a novel framework designed for occluded 3D HPE that capitalizes on the benefits of a discrete diffusion model. Specifically, $\text{Di}^2\text{Pose}$ employs a two-stage process: it first converts 3D poses into a discrete representation through a \emph{pose quantization step}, which is subsequently modeled in latent space through a \emph{discrete diffusion process}. This methodological innovation restrictively confines the search space towards physically viable configurations and enhances the model's capability to comprehend how occlusions affect human pose within the latent space. Extensive evaluations conducted on various benchmarks (e.g., Human3.6M, 3DPW, and 3DPW-Occ) have demonstrated its effectiveness. |
| title | $\text{Di}^2\text{Pose}$: Discrete Diffusion Model for Occluded 3D Human Pose Estimation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2405.17016 |