Cover Image for Canonicalization: Experiments in setting some chairs in order - Prof' Srinath Sridhar

Canonicalization: Experiments in setting some chairs in order - Prof' Srinath Sridhar

Hosted by Peter
 
 
Zoom
Registration
Past Event
Welcome! To join the event, please register below.
You will be asked to verify token ownership with your wallet.
About Event

​[Event was postponed from Feb 21 to March 28 due to restrictions with a recently published paper which is not allowed to be public yet Until March. Apologies all for the inconvenience]

​

​In this talk, I will introduce the notions of invariance, equivariance, and 'object canonicalization' (i.e., mapping object properties to canonical states). I will demonstrate how canonicalization enables us to better solve tasks in computer vision and robotics including 6DoF object pose estimation and 3D reconstruction. I will discuss our work on fully supervised and weakly supervised canonicalization methods, but focus on self-supervised methods. Finally, I will discuss future directions including opportunities for using canonicalization to understand articulated and non-rigid objects.

​The talk is based on the speaker's papers:

  1. ​Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation (CVPR 2019)
    https://geometry.stanford.edu/projects/NOCS_CVPR2019/

  2. ​Multiview Aggregation for Learning Category-Specific Shape Reconstruction (NeurIPS 2019)
    https://geometry.stanford.edu/projects/xnocs/

  3. ​Pix2Surf: Learning Parametric 3D Surface Models of Objects from Images (ECCV 2020)
    https://geometry.stanford.edu/projects/pix2surf/

  4. ​CaSPR: Learning Canonical Spatiotemporal (NeurIPS 2020 Spotlight)
    https://geometry.stanford.edu/projects/caspr/

  5. ​DRACO: Weakly Supervised Dense Reconstruction And Canonicalization of Objects
    https://aadilmehdis.github.io/DRACO-Project-Page/

  6. ​ConDor: Self-Supervised Canonicalization of 3D Pose for Partial Shapes
    https://ivl.cs.brown.edu/ConDor/

​

​Presenter Bio:

​Srinath Sridhar (http://srinathsridhar.com/) is an assistant professor of computer science at Brown University. His research interests are in 3D computer vision and machine learning. Specifically, he focuses on visual understanding of 3D human physical interactions with applications ranging from robotics to mixed reality. He has won several fellowships (e.g., Google Research Scholar) and awards (e.g., Eurographics Best Paper Honorable Mention) for his work, and has previously spent time at Stanford, Max Planck Institute for Informatics, Microsoft Research Redmond, and Honda Research Institute.

​

​------------
Find us at:

Hosts
Peter