SDMNet is a self-supervised stereo matching network trained via image reconstruction and perceptual loss that reports competitive results on KITTI and hand vein datasets without using disparity labels.
Computing visual correspondence with occlusions using graph cuts
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Non-destructive three-dimensional measurement of hand vein based on self-supervised network
SDMNet is a self-supervised stereo matching network trained via image reconstruction and perceptual loss that reports competitive results on KITTI and hand vein datasets without using disparity labels.