GainedVAE

A Pytorch Implementation of a continuously rate adjustable learned image compression framework, Gained Variational Autoencoder(GainedVAE).

Note that This Is Not An Official Implementation Code.

More details can be found in the following paper:

Asymmetric Gained Deep Image Compression With Continuous Rate Adaptation.
Huawei Technologies, CVPR 2021
Ze Cui, Jing Wang, Shangyin Gao, Tiansheng Guo, Yihui Feng, Bo Bai

Todo: Reproduce Implementation of the following paper:

INTERPOLATION VARIABLE RATE IMAGE COMPRESSION
Alibaba Group, arxiv 2021.9.20
Zhenhong Sun, Zhiyu Tan, Xiuyu Sun, Fangyi Zhang, Yichen Qian, Dongyang Li, Hao Li

Environment

• Python == 3.7.10
• Pytorch == 1.7.1
• CompressAI

Dataset

Training set

I use a part of the OpenImages Dataset to train the models (train06, train07, train08, about 54w images). You can download from here. Download OpenImages Maybe train08 (14w images) is enough.

Test set

Download Kodak dataset

Train Your Own Model

python3 trainGain.py -d /path/to/your/image/dataset/ --epochs 200 -lr 1e-4 --batch-size 16 --model-save /path/to/your/model/save/dir --cuda

Result

I try to train the Gained Mean-Scale Hyperprior model and here is the result.

Acknowledgement

The framework is based on CompressAI, I add the model in compressai.models.gain, compressai.models.gain_utils.
And trainGain/trainGain.py is modified with reference to compressai_examples/train.py.

More Variable Rate Image Compression Repositories

"Variable-Rate Deep Image Compression through Spatially-Adaptive Feature Transform" (ICCV 2021).
code

"Variable Bitrate Image Compression with Quality Scaling Factors" (ICASSP 2020).
code

"Variable Rate Deep Image Compression with Modulated Autoencoders" (IEEE SPL 2020)
code

"Slimmable Compressive Autoencoders for Practical Neural Image Compression" (CVPR 2021)
code

Contact

Feel free to contact me if there is any question about the code or to discuss any problems with image and video compression. ([email protected])