Humans in 4D: Reconstructing and Tracking Humans with Transformers
Researchers have developed a new approach to track humans over time called 4DHumans. Using a network called HMR 2.0, the new method shows promising results in reconstructing humans' unusual poses through 3D reconstructions.
![Humans in 4D: Reconstructing and Tracking Humans with Transformers](/content/images/size/w1200/2023/06/Screenshot_20230606_134508.png)
![](https://ssv.ai/content/images/2023/06/Screenshot_20230606_134508-1.png)
Researchers have developed a new approach to track humans over time called 4DHumans. Using a network called HMR 2.0, the new method shows promising results in reconstructing humans' unusual poses through 3D reconstructions. The approach also maintains identities through occlusion events and yields better accuracy in action recognition compared to previous methods. The project code and models are available for further study.
Paper
Humans in 4D: Reconstructing and Tracking Humans with Transformers
We present an approach to reconstruct humans and track them over time. At thecore of our approach, we propose a fully “transformerized” version of a networkfor human mesh recovery. This network, HMR 2.0, advances the state of the artand shows the capability to analyze unusual poses that have in t…
![](https://static.arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png)
Source Code
GitHub - shubham-goel/4D-Humans
Contribute to shubham-goel/4D-Humans development by creating an account on GitHub.