What Do Single-view 3D Reconstruction Networks Learn?

CVPR 2019 β€’ Maxim Tatarchenko β€’ Stephan R. Richter β€’ RenΓ© Ranftl β€’ Zhuwen Li β€’ Vladlen Koltun β€’ Thomas Brox

Convolutional networks for single-view object reconstruction have shown impressive performance and have become a popular subject of research. All existing techniques are united by the idea of having an encoder-decoder network that performs non-trivial reasoning about the 3D structure of the output space... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
3D Reconstruction 300W ResNet 1-of-100 Accuracy cosine loss # 1