3 ms·
This problem has already been solved with decent success using deep learning. See: https://homes.cs.washington.edu/~jxie/pdf/deep3d.pdf https://homes.cs.washin
by nicklo 10y ago
This problem has already been solved with decent success using deep learning.
See: https://homes.cs.washington.edu/~jxie/pdf/deep3d.pdf https://homes.cs.washington.edu/~jxie/pdf/deep3d.pdf
- mrfusion 10y agoThat's a good start. I was thinking you could generate unlimited training data by using a game engine. You'd have the actual 3D model for every single frame.
- phorese 10y agoYup, the community is on it! http://www.cv-foundation.org/openaccess/content_cvpr_2016/html/Ros_The_SYNTHIA_Dataset_CVPR_2016_paper.html http://www.cv-foundation.org/openaccess/content_cvpr_2016/ht... http://www.cv-foundation.org/openaccess/content_cvpr_2016/html/Mayer_A_Large_Dataset_CVPR_2016_paper.html http://www.cv-foundation.org/openaccess/content_cvpr_2016/ht... https://link.springer.com/chapter/10.1007/978-3-319-46475-6_7 https://link.springer.com/chapter/10.1007/978-3-319-46475-6_... And there's more every week... Blender, Unity Engine, Unreal Engine, you name it. (Disclaimer: am author on one of these papers)
- davesque 10y agoI'm aware that certain automotive companies are already doing this.