Self-Supervising Occlusions For Vision | Dinesh Reddy,PhD @CMU Robotics скачать в хорошем качестве

Self-Supervising Occlusions For Vision | Dinesh Reddy,PhD @CMU Robotics

Abstract : Virtually every scene has occlusions. Understanding and dealing with occlusions is hard due to the large variation in the type, number, and extent of occlusions possible in scenes. In this talk, we propose developing computer vision algorithms robust to occlusions using self-supervision. We propose two methodologies for learning such occlusions for data captured in the wild. The first methodology is to predict occluded regions using multi-view supervision. We introduced a large multi-camera dataset capturing activity in the wild for learning Occlusions. In the second segment, we explored using longitudinal data i.e. videos captured over weeks, months, or even years to supervise occluded regions in an object.