Bi3D: Stereo depth estimation via Binary Classification

On February 10th at 12:00, Virtual seminar.

Daniel Corregidor, researcher at the GTI, delivered a virtual seminar on the paper “Bi3D: Stereo depth estimation via Binary Classification" authored by Abhishek Badki, Alejandro Troccoli, Kihwan Kim, Jan Kautz, Pradeep Sen, Orazio Gallo et al. published in Proceedings of the IEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2020, pp. 1600-1608.

Abstract

Stereo-based depth estimation is a cornerstone of computer vision, with state-of-the-art methods delivering accurate results in real time. For several applications such as autonomous navigation, however, it may be useful to trade accuracy for lower latency. We present Bi3D, a method that estimates depth via a series of binary classifications. Rather than testing if objects are at a particular depth D, as existing stereo methods do, it classifies them as being closer or farther than D. This property offers a powerful mechanism to balance accuracy and latency. Given a strict time budget, Bi3D can detect objects closer than a given distance in as little as a few milliseconds, or estimate depth with arbitrarily coarse quantization, with complexity linear with the number of quantization levels. Bi3D can also use the allotted quantization levels to get continuous depth, but in a specific depth range. For standard stereo (i.e., continuous depth on the whole range), our method is close to or on par with state-of-the-art, finely tuned stereo methods.

Tweet