[CVPR 2025] SPIdepth: Strengthened Pose Information for Self-supervised Monocular Depth Estimation

</a> <a href='https://arxiv.org/abs/2404.12501'><img src='https://img.shields.io/badge/Paper-Arxiv-red'></a>

<br> <br> <br>

Training

To train on KITTI, run:

python train.py ./args_files/hisfog/kitti/cvnXt_H_320x1024.txt

For instructions on downloading the KITTI dataset, see Monodepth2

To finetune on KITTI, run:

python ./finetune/train_ft_SQLdepth.py ./conf/cvnXt.txt ./finetune/txt_args/train/inc_kitti.txt

To train on CityScapes, run:

python train.py ./args_files/args_cityscapes_train.txt

To finetune on CityScapes, run:

python train.py ./args_files/args_cityscapes_finetune.txt

For preparing cityscapes dataset, please refer to SfMLearner's prepare_train_data.py script. We used the following command:

python prepare_train_data.py \
    --img_height 512 \
    --img_width 1024 \
    --dataset_dir <path_to_downloaded_cityscapes_data> \
    --dataset_name cityscapes \
    --dump_root <your_preprocessed_cityscapes_path> \
    --seq_length 3 \
    --num_threads 8

Pretrained weights and evaluation

You can download weights for some pretrained models here:

• KITTI
• CityScapes

To evaluate a model on KITTI, run:

python evaluate_depth_config.py args_files/hisfog/kitti/cvnXt_H_320x1024.txt

Make sure you have first run export_gt_depth.py to extract ground truth files.

And to evaluate a model on Cityscapes, run:

python ./tools/evaluate_depth_cityscapes_config.py args_files/args_cvnXt_H_cityscapes_finetune_eval.txt

The ground truth depth files can be found at HERE, Download this and unzip into splits/cityscapes.

Inference with your own images

python test_simple_SQL_config.py ./conf/cvnXt.txt

In ./conf/cvnXt.txt, you can set --image_path to a single image or a directory of images.

Citation

If you find this project useful for your research, please consider citing:

@InProceedings{Lavreniuk_2025_CVPR,
    author    = {Lavreniuk, Mykola and Lavreniuk, Alla},
    title     = {SPIdepth: Strengthened Pose Information for Self-supervised Monocular Depth Estimation},
    booktitle = {Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR) Workshops},
    month     = {June},
    year      = {2025},
    pages     = {874-884}
}

Acknowledgement

This project is built on top of SQLdepth, and we are grateful for their outstanding contributions.