MSFNet fuses multi-scale features with a guidance mechanism and consistency checks to achieve state-of-the-art disparity estimation on Scene Flow and KITTI 2015.
Improved stereo matching with constant highway networks and reflective confidence learning
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
-
End-to-End Learning of Multi-scale Convolutional Neural Network for Stereo Matching
MSFNet fuses multi-scale features with a guidance mechanism and consistency checks to achieve state-of-the-art disparity estimation on Scene Flow and KITTI 2015.