Digging Into Self-Supervised Monocular Depth Estimation

Code

• github.com/nianticlabs/monodepth2
  OfficialIn paperpytorch★ 0
• github.com/FangGet/tf-monodepth2
  tf★ 82
• github.com/XXXVincent/MonoDepth2
  pytorch★ 21
• github.com/minghanz/DepthC3D
  pytorch★ 20
• github.com/tudelft/filled-disparity-monodepth
  tf★ 14
• github.com/qrzyang/pseudo-stereo
  pytorch★ 11
• github.com/rnlee1998/SRD
  pytorch★ 7
• github.com/iodncookie/my_loss_monodepth_master
  tf★ 0
• github.com/IcarusWizard/monodepth2-paddle
  paddle★ 0
• github.com/CaptainEven/MonoDepthV2
  pytorch★ 0

Abstract

Per-pixel ground-truth depth data is challenging to acquire at scale. To overcome this limitation, self-supervised learning has emerged as a promising alternative for training models to perform monocular depth estimation. In this paper, we propose a set of improvements, which together result in both quantitatively and qualitatively improved depth maps compared to competing self-supervised methods. Research on self-supervised monocular training usually explores increasingly complex architectures, loss functions, and image formation models, all of which have recently helped to close the gap with fully-supervised methods. We show that a surprisingly simple model, and associated design choices, lead to superior predictions. In particular, we propose (i) a minimum reprojection loss, designed to robustly handle occlusions, (ii) a full-resolution multi-scale sampling method that reduces visual artifacts, and (iii) an auto-masking loss to ignore training pixels that violate camera motion assumptions. We demonstrate the effectiveness of each component in isolation, and show high quality, state-of-the-art results on the KITTI benchmark.

Tasks

Benchmark Results

Reproductions