Monocular Depth Prediction
[NEW] Training codes have been uploaded!
This repository contains the source code of our paper: Yin Wei, Yifan Liu, Chunhua Shen, Youliang Yan, Enforcing geometric constraints of virtual normal for depth prediction (accepted for publication in ICCV' 2019).
NYUDV2 The details of datasets can be found here. The Eigen split of labeled images can be downloaded here. Please extract it to ./datasets. Our SOTA model is trained on the around 20K unlabled images.
KITTI The details of KITTI benchmark for monocular depth prediction is here. We use both the official split and Eigen split. You can find the filenames here.
# Run the inferece on NYUDV2 dataset
python ./tools/test_nyu_metric.py \
--dataroot ./datasets/NYUDV2 \
--dataset nyudv2 \
--cfg_file lib/configs/resnext101_32x4d_nyudv2_class \
--load_ckpt ./nyu_rawdata.pth
# Test depth predictions on any images, please replace the data dir in test_any_images.py
python ./tools/test_any_images.py \
--dataroot ./ \
--dataset any \
--cfg_file lib/configs/resnext101_32x4d_nyudv2_class \
--load_ckpt ./nyu_rawdata.pth
If you want to test the kitti dataset, please see here
# Run the inferece on NYUDV2 dataset
python ./tools/train_nyu_metric.py \
--dataroot ./datasets/NYUDV2 \
--dataset nyudv2 \
--cfg_file lib/configs/resnext101_32x4d_nyudv2_class
#--load_ckpt *.pth
#--resume
```
### Citation
@inproceedings{Yin2019enforcing, title={Enforcing geometric constraints of virtual normal for depth prediction}, author={Yin, Wei and Liu, Yifan and Shen, Chunhua and Yan, Youliang}, booktitle= {The IEEE International Conference on Computer Vision (ICCV)}, year={2019} }
@article{yin2021virtual, title={Virtual Normal: Enforcing Geometric Constraints for Accurate and Robust Depth Prediction}, author={Yin, Wei and Liu, Yifan and Shen, Chunhua}, journal={IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)}, year={2021} }
## Other works :)
We have done a new work to aim to solve the generalization problem of monocular depth prediction. Please refer to following link. :)
- Paper: DiverseDepth: Affine-invariant Depth Prediction Using Diverse Data
- Code: https://github.com/YvanYin/DiverseDepth
### Contact
Wei Yin: [email protected]