2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2017)
Honolulu, Hawaii, USA
July 21, 2017 to July 26, 2017
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CVPR.2017.596
In this paper we formulate structure from motion as a learning problem. We train a convolutional network end-to-end to compute depth and camera motion from successive, unconstrained image pairs. The architecture is composed of multiple stacked encoder-decoder networks, the core part being an iterative network that is able to improve its own predictions. The network estimates not only depth and motion, but additionally surface normals, optical flow between the images and confidence of the matching. A crucial component of the approach is a training loss based on spatial relative differences. Compared to traditional two-frame structure from motion methods, results are more accurate and more robust. In contrast to the popular depth-from-single-image networks, DeMoN learns the concept of matching and, thus, better generalizes to structures not seen during training.
cameras, image matching, image motion analysis, image sequences, iterative methods, learning (artificial intelligence), stereo image processing
B. Ummenhofer et al., "DeMoN: Depth and Motion Network for Learning Monocular Stereo," 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii, USA, 2017, pp. 5622-5631.