Title | ||
---|---|---|
You2Me - Inferring Body Pose in Egocentric Video via First and Second Person Interactions. |
Abstract | ||
---|---|---|
The body pose of a person wearing a camera is of great interest for applications in augmented reality, healthcare, and robotics, yet much of the person\u0027s body is out of view for a typical wearable camera. We propose a learning-based approach to estimate the camera wearer\u0027s 3D body pose from egocentric video sequences. Our key insight is to leverage interactions with another person---whose body pose we can directly observe---as a signal inherently linked to the body pose of the first-person subject. We show that since interactions between individuals often induce a well-ordered series of back-and-forth responses, it is possible to learn a temporal model of the interlinked poses even though one party is largely out of view. We demonstrate our idea on a variety of domains with dyadic interaction and show the substantial impact on egocentric body pose estimation, which improves the state of the art. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/CVPR42600.2020.00991 | CVPR |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
43 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Evonne Ng | 1 | 0 | 0.68 |
donglai xiang | 2 | 67 | 4.14 |
Joo Hanbyul | 3 | 2 | 2.38 |
Kristen Grauman | 4 | 6258 | 326.34 |