Probabilistic object detection: Strengths, Weaknesses, and Opportunities

Authors: Dhaivat Bhatt*, Dishank Bansal*, Gunshi Gupta*, Hanju Lee, Krishna Murthy Jatavallabhula, Liam Paull

Paper can be found here.

Deep neural networks are the de-facto standard for object detection in autonomous driving applications. However, neural networks cannot be blindly trusted even within the training data distribution, let alone outside it. This has paved way for several probabilistic object detection techniques that measure uncertainty in the outputs of an object detector. Through this position paper, we serve three main purposes. First, we briefly sketch the landscape of current methods for probabilistic object detection. Second, we present the main shortcomings of these approaches. Finally, we present promising avenues for future research, and proof-of-concept results where applicable. Through this effort, we hope to bring the community one step closer to performing accurate, reliable, and consistent probabilistic object detection.

Presented at the ICML AIAD 2020 Workshop

We survey and describe the strengths of current probabilistic deep learning methods and how they are employed for object detection. This is followed by a critique of the major weaknesses of these methods, and we conclude by analyzing the research opportunities that open up.