3DB: A Framework for Debugging Computer Vision Models
Guillaume Leclerc
Hadi Salman
Andrew Ilyas
Sai H. Vemprala
Logan Engstrom
Vibhav Vineet
Kai Y. Xiao
Pengchuan Zhang
Shibani Santurkar
Greg Yang
Ashish Kapoor
A. Madry

Abstract
We introduce 3DB: an extendable, unified framework for testing and debugging vision models using photorealistic simulation. We demonstrate, through a wide range of use cases, that 3DB allows users to discover vulnerabilities in computer vision systems and gain insights into how models make decisions. 3DB captures and generalizes many robustness analyses from prior work, and enables one to study their interplay. Finally, we find that the insights generated by the system transfer to the physical world. We are releasing 3DB as a library (https://github.com/3db/3db) alongside a set of example analyses, guides, and documentation: https://3db.github.io/3db/ .
View on arXivComments on this paper