Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/KexianHust/Structure-Guided-Ranking-Loss
Structure-Guided Ranking Loss for Single Image Depth Prediction
https://github.com/KexianHust/Structure-Guided-Ranking-Loss
cvpr2020 monocular-depth-estimation
Last synced: 2 months ago
JSON representation
Structure-Guided Ranking Loss for Single Image Depth Prediction
- Host: GitHub
- URL: https://github.com/KexianHust/Structure-Guided-Ranking-Loss
- Owner: KexianHust
- License: other
- Created: 2020-03-12T02:25:42.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2023-07-19T11:23:00.000Z (11 months ago)
- Last Synced: 2024-01-21T05:00:05.309Z (5 months ago)
- Topics: cvpr2020, monocular-depth-estimation
- Language: Python
- Size: 16.7 MB
- Stars: 174
- Watchers: 18
- Forks: 16
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
Lists
- awesome-rgbd-datasets - HRWSI - the-wild |Color, Depth | |20778 images |2020 | (RGB-D Datasets <a id="list" class="anchor" href="#list" aria-hidden="true"><span class="octicon octicon-link"></span></a>)
README
# Structure-Guided Ranking Loss for Single Image Depth Prediction
This repository contains a pytorch implementation of our CVPR2020 paper "Structure-Guided Ranking Loss for Single Image Depth Prediction".
[Project Page](https://KexianHust.github.io/Structure-Guided-Ranking-Loss/)
## Changelog
* [Jun. 2020] Initial release
## To do
- [ ] Mix data training
## Prerequisites
* Pytorch >= 0.4.1
* CUDA >= 0.8
* Python >= 2.7
* glob, matplotlib
* Need to compile the syncbn module in models/syncbn. Note that the directory of the syncbn module should be modified in some .py files (i.e., DepthNet.py, resnet.py and networks.py)
* Download the [model.pth.tar](https://drive.google.com/file/d/1p8c8-nUTNry5usQmGdTC2TrwWrp3dQ0y/view?usp=sharing)
## Inference
```bash
# Before running, you should set the CUDA_VISIBLE_DEVICES in demo.sh
bash demo.sh
```
If you find our work useful in your research, please consider citing the paper.
```
@InProceedings{Xian_2020_CVPR,
author = {Xian, Ke and Zhang, Jianming and Wang, Oliver and Mai, Long and Lin, Zhe and Cao, Zhiguo},
title = {Structure-Guided Ranking Loss for Single Image Depth Prediction},
booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2020}
}
```
## Dataset
Our [HRWSI](https://drive.google.com/file/d/1OVOx6x-B0Cs-m2z_-7ZxSgRFHz_VBvDd/view?usp=sharing) dataset is for research only! Some researchers may interested in the stereo data, so we provide the right views [here](https://drive.google.com/file/d/1HzEB7yQI05Q21dP9rRjnyMoEmvCckAQp/view?usp=sharing). Please let me know if you have any questions.
## Lisence
Research only