NeurIPS 2020

Targeted Adversarial Perturbations for Monocular Depth Prediction

Meta Review

All reviewers appreciate the findings presented in this paper regarding the robustness of depth estimation networks. All reviewers recommend accept after the rebuttal. AC agrees with this consensus recommendation. The authors successfully addressed the reviewers' concerns in the rebuttal. The authors should revise the paper by incorporating the clarifications and responses presented in the rebuttal, in particular, adding the additional quantitative results and the additional discussion about transferability. NOTE FROM PROGRAM CHAIRS: For the camera-ready version, please expand your broader impact statement to include a more substantive discussion of the potential negative impacts of your work.