# DynaFill
[Project](http://inpainting.cs.uni-freiburg.de/) | [arXiv](https://arxiv.org/abs/2008.05058)
Borna Bešić, Abhinav Valada
_Dynamic Object Removal and Spatio-Temporal RGB-D Inpainting via Geometry-Aware Adversarial Learning_
![](img/slider.gif)
## Setting Up the Environment
We recommend using [conda](https://docs.conda.io/en/latest/) package and environment management system. We provide [`environment.yml`](environment.yml) that can be used to easily create a self-contained environment with all the dependencies:
```sh
conda env create -f environment.yml
conda activate DynaFill
```
#### Tested Configuration
- Linux 4.15.0-122-generic x86_64
- NVIDIA GPU Driver 390.138 + CUDA 9.0
- Python 3.6
- PyTorch 1.1.0 + torchvision 0.2.1
- OpenCV 4.0
## Dataset
The description of our DynaFill dataset with the corresponding download instructions can be found at [inpainting.cs.uni-freiburg.de/#dataset](http://inpainting.cs.uni-freiburg.de/#dataset).
## Running the Demo
```sh
usage: demo.py [-h] [--device DEVICE] dataset_split_dir
positional arguments:
dataset_split_dir Path to training/ or validation/ directory of DynaFill
dataset
optional arguments:
-h, --help show this help message and exit
--device DEVICE Device on which to run inference
```
#### Example
```sh
python demo.py /mnt/data/DynaFill/validation --device cuda:1
```
## Citation
If you find the code useful for your research, please consider citing our paper:
```
@article{bei2020dynamic,
title={Dynamic Object Removal and Spatio-Temporal RGB-D Inpainting via Geometry-Aware Adversarial Learning},
author={Borna Bešić and Abhinav Valada},
journal={arXiv preprint arXiv:2008.05058},
year={2020}
}
```
## License
For academic usage, the code is released under the [GPLv3 license](LICENSE). For any commercial purpose, please contact the authors.
没有合适的资源?快使用搜索试试~ 我知道了~
通过几何感知对抗学习进行动态对象移除和时空RGB-D修复.zip
共24个文件
py:16个
pth:3个
gif:1个
1.该资源内容由用户上传,如若侵权请联系客服进行举报
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
版权申诉
0 下载量 95 浏览量
2023-03-22
21:52:43
上传
评论
收藏 79.03MB ZIP 举报
温馨提示
通过几何感知对抗学习进行动态对象移除和时空RGB-D修复
资源推荐
资源详情
资源评论
收起资源包目录
通过几何感知对抗学习进行动态对象移除和时空RGB-D修复.zip (24个子文件)
DynaFill-master
utils.py 547B
LICENSE 34KB
img
slider.gif 345KB
environment.yml 232B
modules.py 2KB
models
__init__.py 0B
gan
__init__.py 0B
ours
__init__.py 0B
refinement_resnet.py 2KB
coarse_resnet.py 1009B
blocks_vanilla.py 8KB
discriminator.py 1012B
depth
self_supervised_sparse_to_dense
__init__.py 131B
model.py 7KB
criteria.py 3KB
functional.py 394B
checkpoints
refined_translation.pth 17.12MB
depth_completion.pth 61.11MB
coarse_inpainting.pth 6.64MB
datasets.py 4KB
.gitignore 13B
demo.py 7KB
README.md 2KB
flow.py 8KB
共 24 条
- 1
资源评论
快撑死的鱼
- 粉丝: 1w+
- 资源: 9154
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功