2 Hits in 1.5 sec

Bi3D: Stereo Depth Estimation via Binary Classifications [article]

Abhishek Badki, Alejandro Troccoli, Kihwan Kim, Jan Kautz, Pradeep Sen, Orazio Gallo
2020 arXiv   pre-print
We present Bi3D, a method that estimates depth via a series of binary classifications.  ...  Stereo-based depth estimation is a cornerstone of computer vision, with state-of-the-art methods delivering accurate results in real time.  ...  Bi3D: Stereo Depth Estimation via Binary Classifications (Supplementary) Additional Details on Training FeatNet.  ... 
arXiv:2005.07274v2 fatcat:acyx2jh5jbgx3khrx6ocu42rae

Revisiting Stereo Depth Estimation From a Sequence-to-Sequence Perspective with Transformers [article]

Zhaoshuo Li, Xingtong Liu, Nathan Drenkow, Andy Ding, Francis X. Creighton, Russell H. Taylor, Mathias Unberath
2021 arXiv   pre-print
Stereo depth estimation relies on optimal correspondence matching between pixels on epipolar lines in the left and right images to infer depth.  ...  This approach, named STereo TRansformer (STTR), has several advantages: It 1) relaxes the limitation of a fixed disparity range, 2) identifies occluded regions and provides confidence estimates, and 3)  ...  Related Work Stereo Depth Estimation In general, the stereo depth estimation task involves two key steps (1) feature matching and (2) matching cost aggregation [23] .  ... 
arXiv:2011.02910v4 fatcat:q46pc3w2zzbnzcapmq5mpsrkaa