Instructions:

• Install Pycharm: there is a free version. Good for debugging.
• Install python 3.5 (I have 3.5.2) or 3.6 + libraries + Tensorflow gpu (I have 1.8.0)
• Python dependencies: Tensorlayer, numpy(1.16.1), scipy, matplotlib, sklearn and imageio

Usage

Run YUV model evaluation

--mode=evaluate --evalHR=$output yuv path$ --evalLR=$input yuv path$ --testModel=$MFRNet_BVI_VTM70_SR_QPXX.npz$ --ratio=1 --nlayers=16 --GAN=0 --nframes=0 --eval_inputType=YUV --readBatch_flag=1 --inputFormat=RGB

All input files should use the standard file name, e.g. Campfire_3840x2160_30fps_10bit_qp22.yuv (filename_HxW_xxxfps_xbit_qpxx.yuv)

Reference

Please cite our papers if you use this code for your research

[1] Ma, Di, Fan Zhang, and David R. Bull. "MFRNet: a new CNN architecture for post-processing and in-loop filtering." IEEE Journal of Selected Topics in Signal Processing 15.2 (2020): 378-387. [2] Feng, Chen, et al. "Enhancing HDR Video Compression through CNN-based Effective Bit Depth Adaptation." arXiv preprint arXiv:2207.08634 (2022).