basic.md

Basic information about depth estimation

Datasets

1. KITTI
2. NYUD
3. Cityscapes
4. Make3D

Evaluation Metrics

1. RMSE
2. RMSE log
3. Abs Rel
4. Sq Rel
5. Accuracies

Losses

1. L2 loss.
2. The one proposed in Depth map prediction from a single image using a multi-scale deep network (Eigen).
3. Predicting depth, surface normals and semantic labels with a common multi-scale convolutional architecture (Eigen)
4. Berhu Loss.
5. Disparity L2 loss.