Weakly-Supervised Discovery of Geometry-Aware Representation for 3D Human Pose Estimation Article Swipe

MAKE IMPACT

Xipeng Chen , Kwan-Yee Lin , Wentao Liu , Chen Qian , Xiaogang Wang , Liang Lin · YOU? · · 2019 · Open Access · · DOI: https://doi.org/10.48550/arxiv.1903.08839

Recent studies have shown remarkable advances in 3D human pose estimation from monocular images, with the help of large-scale in-door 3D datasets and sophisticated network architectures. However, the generalizability to different environments remains an elusive goal. In this work, we propose a geometry-aware 3D representation for the human pose to address this limitation by using multiple views in a simple auto-encoder model at the training stage and only 2D keypoint information as supervision. A view synthesis framework is proposed to learn the shared 3D representation between viewpoints with synthesizing the human pose from one viewpoint to the other one. Instead of performing a direct transfer in the raw image-level, we propose a skeleton-based encoder-decoder mechanism to distil only pose-related representation in the latent space. A learning-based representation consistency constraint is further introduced to facilitate the robustness of latent 3D representation. Since the learnt representation encodes 3D geometry information, mapping it to 3D pose will be much easier than conventional frameworks that use an image or 2D coordinates as the input of 3D pose estimator. We demonstrate our approach on the task of 3D human pose estimation. Comprehensive experiments on three popular benchmarks show that our model can significantly improve the performance of state-of-the-art methods with simply injecting the representation as a robust 3D prior.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/1903.08839

PDF

https://arxiv.org/pdf/1903.08839

OA Status

green

Cited By

42

References

43

Related Works

10

OpenAlex ID

https://openalex.org/W2924460655

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W2924460655

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.1903.08839

Digital Object Identifier

Title

Weakly-Supervised Discovery of Geometry-Aware Representation for 3D Human Pose Estimation

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2019

Year of publication

Publication date

2019-03-21

Full publication date if available

Authors

Xipeng Chen, Kwan-Yee Lin, Wentao Liu, Chen Qian, Xiaogang Wang, Liang Lin

List of authors in order

Landing page

https://arxiv.org/abs/1903.08839

Publisher landing page

PDF URL

https://arxiv.org/pdf/1903.08839

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/1903.08839

Direct OA link when available

Concepts

Pose, Computer science, Artificial intelligence, Representation (politics), Robustness (evolution), Encoder, Computer vision, Feature learning, 3D pose estimation, Estimator, Leverage (statistics), Machine learning, Pattern recognition (psychology), Mathematics, Gene, Biochemistry, Law, Statistics, Political science, Chemistry, Operating system, Politics

Top concepts (fields/topics) attached by OpenAlex

Cited by

42

Total citation count in OpenAlex

Citations by year (recent)

2023: 3, 2022: 5, 2021: 13, 2020: 17, 2019: 4

Per-year citation counts (last 5 years)

References (count)

43

Number of works referenced by this work

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload