Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
CVPR2023 (highlight) - A Dynamic Multi-Scale Voxel Flow Network for Video Prediction
| Date | Stars |
|---|---|
| 2026-07-24 | 368 |
| 2026-07-25 | 368 |
| 2026-07-28 | 368 |
| 2026-07-30 | 368 |
| 2026-08-06 | 368 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# A Dynamic Multi-Scale Voxel Flow Network for Video Prediction
## [HomePage](https://huxiaotaostasy.github.io/DMVFN) | [Colab](https://colab.research.google.com/github/megvii-research/CVPR2023-DMVFN/blob/main/colab_demo.ipynb) | [arXiv](https://arxiv.org/abs/2303.09875) | [YouTube](https://youtu.be/6retLBLAHDs)
This project is the implementation of our Paper: [A Dynamic Multi-Scale Voxel Flow Network for Video Prediction](https://arxiv.org/abs/2303.09875), which is accepted by **CVPR2023 (highlight✨, 10% of accepted papers).** We proposed a SOTA model for Video Prediction.
[Poster](https://drive.google.com/file/d/1W5MD39Wjp4Ryrv_DMC7QXiCNsiG1AL2x/view?usp=sharing) | [研究历程](https://www.zhihu.com/question/585474435/answer/2946859435) | [中文论文](https://drive.google.com/file/d/1mzJipcWRd1wcpphSIUOszzgxMWMFx40D/view?usp=sharing) | [rebuttal (3WA->1WA2SA)](https://drive.google.com/file/d/1S8tZOokoZNxCBkV1GVLWOlUCCoJA7Bfu/view?usp=sharing) | [Demo](https://youtu.be/rlghCGbAqUo)
supplement and correction of published paper: 1. In Figure 2(a), $I_{t-1}$ and $I_t$ should be swapped; 2. "Our model is trained on four 2080Ti GPUs for 300 epochs, which takes about 35 hours." This sentence refers to the Cityscapes training set; 3. In Table 1, GFLOPs computing for KITTI is on $256\times832$ and on $1024\times512$ for Cityscapes using [fvcore](https://github.com/facebookresearch/fvcore).
## Usage
### Installation
```bash
git clone https://github.com/megvii-research/CVPR2023-DMVFN.git
cd CVPR2023-DMVFN
pip3 install -r requirements.txt
```
* Download the pretrained models from [Google Drive](https://drive.google.com/drive/folders/1QnsnTVU-2IexFNjGhNVRYPqKNviAbkwH?usp=sharing). ([百度网盘](https://pan.baidu.com/s/1l8d1oCzEHDf8JCv1KLIbvA) password:33ly), and move the pretrained parameters to `CVPR2023-DMVFN/pretrained_models/*`
```bash
pip install gdown
mkdir pretrained_models && cd pretrained_models
gdown --id 1jILbS8Gm4E5Xx4tDCPZh_7rId0eo8r9W
gdown --id 1WrV30prRiS4hWOQBnVPUxdaTlp9XxmVK
gdown --id 14_xQ3Yl3mO89hr28hbcQW3h63lLrcYY0
cd ..
```
### Data Preparation
In this section, we will download all parts including training and testing sets. If you only need the test set, please jump to [Directly download test splits](#directly_download_test_splits).
The final data folder `CVPR2023-DMVFN/data/` should be organized like this:
```
data
├── Cityscapes
│ ├── train (Citysapes images in 512x1024)
│ │ └── 000000
│ │ └── ...
│ │ └── 002974
│ ├── test (Citysapes images in 512x1024)
│ │ └── 000000
│ │ └── ...
│ │ └── 000499
├── KITTI
│ ├── train (Kitti images in 256x832)
│ │ └── 000000
│ │ └── ...
│ │ └── 013499
│ ├── test (Kitti images in 256x832)
│ │ └── 000000
│ │ └── ...
│ │ └── 001336
├── UCF101
│ ├── v_ApplyEyeMakeup_g08_c01
│ └── ...
└── vimeo_interp_test
└── target
└── 00001
└── ...
```
**Cityscapes**
* Download the Cityscapes dataset `leftImg8bit_sequence_trainvaltest.zip` from [here](https://www.cityscapes-dataset.com/downloads/).
* Unzip `leftImg8bit_sequence_trainvaltest.zip`.
```bash
unzip leftImg8bit_sequence_trainvaltest.zip
```
* Run ./utils/prepare_city.py
```bash
python3 ./utils/prepare_city.py
```
**KITTI**
* Download the KITTI dataset from [Google Drive](https://www.cvlibs.net/datasets/kitti/raw_data.php). You need to register and login, then download all videos.
* Our training split and testing split are consistent with [YueWuHKUST/CVPR2020-FutureVideoSynthesis](https://github.com/YueWuHKUST/CVPR2020-FutureVideoSynthesis).
```
Videos for training split:
2011_09_26_drive_0001_sync 2011_09_26_drive_0018_sync 2011_09_26_drive_0104_sync 2011_09_26_drive_0002_sync 2011_09_26_drive_0048_sync
2011_09_26_drive_0106_sync 2011_09_26_drive_0005_sync 2011_09_26_drive_0051_sync 2011_09_26_drive_0113_sync 2011_09_26_drive_0009_sync
2011_09_26_drive_0056_sync 2011_09_26_drive_0117_sync 2011_09_26_Excerpt of 9,227 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:bd7916c247aaa636, topic:deep-learning, topic:pytorch
matched fp:bd7916c247aaa636, topic:computer-vision