Summary of Whac: World-grounded Humans and Cameras, by Wanqi Yin et al.
WHAC: World-grounded Humans and Cameras
by Wanqi Yin, Zhongang Cai, Ruisi Wang, Fanzhou Wang, Chen Wei, Haiyi Mei, Weiye Xiao, Zhitao Yang, Qingping Sun, Atsushi Yamashita, Ziwei Liu, Lei Yang
First submitted to arxiv on: 19 Mar 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Graphics (cs.GR); Machine Learning (cs.LG); Robotics (cs.RO)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel approach to estimate human and camera trajectories from monocular videos, focusing on recovering accurate scale in the world coordinate system. The method, called WHAC, integrates insights from camera-frame SMPL-X estimation methods, which readily recover absolute human depth, and human motions that inherently provide absolute spatial cues. The framework jointly estimates expressive parametric human models (SMPL-X) and corresponding camera poses without relying on traditional optimization techniques. To evaluate the approach, a new synthetic dataset, WHAC-A-Mole, is introduced, featuring diverse interactive human motions and realistic camera trajectories. Extensive experiments on both standard and newly established benchmarks demonstrate the superiority and efficacy of the proposed framework. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper tries to figure out where people are in the world from just a video taken with one camera. It’s a tough problem because we can’t see everything at once. The researchers came up with a new way to solve this problem, using two important clues: what people do and how cameras move. They created a special method called WHAC that helps us figure out where people are in the world and how they’re moving, as well as where the camera is pointing. To test their idea, they made a fake dataset with lots of different scenes and actions. The results show that their approach works really well. |
Keywords
* Artificial intelligence * Optimization
