https://github.com/zjcv/c3d
[ICCV 2015] Learning Spatiotemporal Features with 3D Convolutional Networks
https://github.com/zjcv/c3d
3d-convolutional-networks c3d pytorch
Last synced: 25 days ago
JSON representation
[ICCV 2015] Learning Spatiotemporal Features with 3D Convolutional Networks
- Host: GitHub
- URL: https://github.com/zjcv/c3d
- Owner: ZJCV
- License: apache-2.0
- Created: 2020-08-27T15:02:10.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2020-09-12T13:01:42.000Z (over 5 years ago)
- Last Synced: 2025-10-20T08:31:55.052Z (8 months ago)
- Topics: 3d-convolutional-networks, c3d, pytorch
- Language: Python
- Homepage:
- Size: 45.9 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
«C3D» 复现了论文Learning Spatiotemporal Features with 3D Convolutional Networks
提出的视频分类模型
`C3D`扩展了卷积核的维度，通过加入时间维度，将卷积核从空间感受野（`HxW`）扩展到时空感受野（`TxHxW`），能够有效的捕捉视频片段中的动作信息
## 内容列表
- [内容列表](#内容列表)
- [使用](#使用)
- [主要维护人员](#主要维护人员)
- [参与贡献方式](#参与贡献方式)
- [许可证](#许可证)
## 使用
训练命令如下：
```
$ export CUDA_VISIBLE_DEVICES=1
$ export PYTHONPATH=.
$ python tools/train.py --config_file configs/c3d_hmdb51.yaml
```
*本工程使用了`PyTorch`实现的`HMDB51`和`UCF101`数据集类，其解析和加载速度非常慢*
## 主要维护人员
* zhujian - *Initial work* - [zjykzj](https://github.com/zjykzj)
## 参与贡献方式
欢迎任何人的参与！打开[issue](https://github.com/zjykzj/C3D/issues)或提交合并请求。
注意:
* `GIT`提交，请遵守[Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0-beta.4/)规范
* 语义版本化，请遵守[Semantic Versioning 2.0.0](https://semver.org)规范
* `README`编写，请遵守[standard-readme](https://github.com/RichardLitt/standard-readme)规范
## 许可证
[Apache License 2.0](LICENSE) © 2020 zjykzj