Overview

This is the model that implemented on DeepVNC. This model is improved from A Unified End-to-End Framework for Efficient Deep Image Compression.

Content

• Prerequisites
• Data Preparation
• Training

Prerequisites

You should install the libraries of this repo.

pip install -r requirements.txt

Data Preparation

We need to first prepare the training and validation data. The trainging data is from flicker.com. You can obtain the training data according to description of CompressionData.

The validation data is the popular kodak dataset.

bash data/download_kodak.sh

Training

For high bitrate (1024, 2048, 4096), the out_channel_N is 192 and the out_channel_M is 320 in 'config_high.json'. For low bitrate (128, 256, 512), the out_channel_N is 128 and the out_channel_M is 192 in 'config_low.json'.

For high bitrate of 4096, we first train from scratch as follows.

CUDA_VISIBLE_DEVICES=0 python train_improve_cbam_ssim.py --config examples/example/config_high.json -n baseline_8192 --train flicker_path --val kodak_path

For other high bitrate (1024, 2048), we use the converged model of 8192 as pretrain model and set the learning rate as 1e-5. The training iterations are set as 500000.

The low bitrate (128, 256, 512) training process follows the same strategy.

Author

Wang Hejun@ICA/CAEP