RI Seminar: Shubham Tulsiani : Towards Reconstructing Any Object in 3 D
Shubham Tulsiani Assistant Professor Robotics Institute, Carnegie Mellon University October 22, 2021 Towards Reconstructing Any Object in 3D Abstract: The world we live in is incredibly diverse, comprising of over 10k natural and manmade object categories. While the computer vision community has made impressive progress in classifying images from such diverse categories, the stateoftheart 3D prediction systems are still limited to merely tens of object classes. A key reason for this stark difference is the relative difficulty of acquiring supervision while it is easy to annotate a semantic label for an image, obtaining groundtruth 3D for learning at scale is infeasible. But do we really need such groundtruth 3D for learning In this talk, I will present a learningbased approach that can train from unstructured image collections, using only segmentation outputs from offtheshelf recognition syst
|
|