Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/interdigitalinc/compressai-vision
CompressAI-Vision helps you design, test and compare Video Compression for Machines pipelines. Compression methods can be either pulled from custom AI-based modules from CompressAI or traditional codecs such as H.266/VVC.
https://github.com/interdigitalinc/compressai-vision
compression computer-vision deep-learning machine-to-machine-communication pytorch video-compression
Last synced: about 14 hours ago
JSON representation
CompressAI-Vision helps you design, test and compare Video Compression for Machines pipelines. Compression methods can be either pulled from custom AI-based modules from CompressAI or traditional codecs such as H.266/VVC.
- Host: GitHub
- URL: https://github.com/interdigitalinc/compressai-vision
- Owner: InterDigitalInc
- License: bsd-3-clause-clear
- Created: 2022-09-14T21:56:37.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2024-11-13T17:25:32.000Z (1 day ago)
- Last Synced: 2024-11-13T18:28:24.477Z (1 day ago)
- Topics: compression, computer-vision, deep-learning, machine-to-machine-communication, pytorch, video-compression
- Language: Python
- Homepage: https://interdigitalinc.github.io/CompressAI-Vision/
- Size: 63.9 MB
- Stars: 86
- Watchers: 3
- Forks: 5
- Open Issues: 0
-
Metadata Files:
- Readme: README.MD
- Changelog: NEWS.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
CompressAI-Vision helps you to develop, test and evaluate compression models with standardized tests in the context of compression methods optimized for machine tasks algorithms such as Neural-Network (NN)-based detectors.
It currently focuses on two types of pipeline:
- Video compression for remote inference (`compressai-remote-inference`), which corresponds to the MPEG "Video Coding for Machines" (VCM) activity.
- Split inference (`compressai-split-inference`), which includes an evaluation framework for compressing intermediate features produced in the context of split models. The software supports all thepipelines considered in the related MPEG activity: "Feature Compression for Machines" (FCM).
## Features
- [Detectron2](https://detectron2.readthedocs.io/en/latest/index.html) is used for object detection (Faster-RCNN) and instance segmentation (Mask-RCNN)
- [JDE](https://github.com/Zhongdao/Towards-Realtime-MOT) is used for Object Tracking
## Documentation
A complete documentation is provided [here](https://interdigitalinc.github.io/CompressAI-Vision/index.html), including [installation](https://interdigitalinc.github.io/CompressAI-Vision/installation), [CLI usage](https://interdigitalinc.github.io/CompressAI-Vision/cli_usage.html), as well as [tutorials](https://interdigitalinc.github.io/CompressAI-Vision/tutorials).
## installation
### initialization of the environment
To get started locally and install the development version of CompressAI-Vision, first create a [virtual environment](https://docs.python.org/3.8/library/venv.html) with python==3.8:
```
python3.8 -m venv venv
source ./venv/bin/activate
pip install -U pip
```
The CompressAI library providing learned compresion modules is available as a submodule. It can be initilized by running:
```
git submodule update --init --recursive
```
Note: the installation script documented below installs compressai from source expects the submodule to be populated.
### installation of compressai-vision and supported vision models:
First, if you want to manually export CUDA related paths, please source (e.g. for CUDA 11.8):
```
bash scripts/env_cuda.sh 11.8
```
Then, run:, please run:
```
bash scripts/install.sh
```
For more otions, check:
```
bash scripts/install.sh --help
```
NOTE 1: install.sh gives you the possibility to install vision models' source and weights at specified locations so that mutliple versions of compressai-vision can point to the same installed vision models
NOTE 2: the downlading of JDE pretrained weights might fail. Check that the size of following file is ~558MB.
path/to/weights/jde/jde.1088x608.uncertainty.pt
The file can be downloaded at the following link (in place of the above file path):
"https://docs.google.com/uc?export=download&id=1nlnuYfGNuHWZztQHXwVZSL_FvfE551pA"
## Usage
### Split inference pipelines
To run split-inference pipelines, please use the following command:
```
compressai-split-inference --help
```
Note that the following entry point is kept for backward compability. It runs split inference as well.
```
compressai-vision-eval --help
```
For example for testing a full split inference pipelines without any compression, run
```
compressai-vision-eval --config-name=eval_split_inference_example
```
### Remote inference pipelines
For remote inference (MPEG VCM-like) pipelines, please run:
```
compressai-remote-inference --help
```
### Configurations
Please check other configuration examples provided in ./cfgs as well as examplary scripts in ./scripts
Test data related to the MPEG FCM activity can be found in ./data/mpeg-fcm/
## For developers
After your dev, you can run (and adapt) test scripts from the scripts/tests directory. Please check scripts/tests/Readme.md for more details
### Contributing
Code is formatted using black and isort. To format code, type:
```
make code-format
```
Static checks with those same code formatters can be run manually with:
```
make static-analysis
```
### Compiling documentation
To produce the html documentation, from [docs/](docs/), run:
```
make html
```
To check the pages locally, open [docs/_build/html/index.html](docs/index.html)
## License
CompressAI-Vision is licensed under the BSD 3-Clause Clear License
## Authors
Fabien Racapé, Hyomin Choi, Eimran Eimon, Sampsa Riikonen, Jacky Yat-Hong Lam
## Related links
* [HEVC HM reference software](https://hevc.hhi.fraunhofer.de)
* [VVC VTM reference software](https://vcgit.hhi.fraunhofer.de/jvet/VVCSoftware_VTM)
* [Detectron2](https://detectron2.readthedocs.io/en/latest/index.html)
* [JDE](https://github.com/Zhongdao/Towards-Realtime-MOT.git)