Motion compensation

[RenYang-home]

Dear authors,

It is a great efficient network of deep video compression. It seems that there is no motion compensation network and it only uses warping to compensate motion. Will this lead to a drop in performance? Could you please provide some experimental results of this code?

Thanks,
Ren

[prasangadhungel]

Dear Ren,

Yes, the doubt that the removal of motion compensation net will lead to drop in performance, is perfectly reasonable. Some experiments performed on Vimeo-90k Dataset (separated as test sequence) have shown similar result or a slight drop in performance compared to DVC but with much gain in terms of speed and memory usage. We have to validate the results in other standard test datasets of different resolutions. As it turned out, general test sequences are not publicly available, we have contacted the authors for the datasets. As soon as we get the datasets and carry out the evaluation we will provide the result in this repo itself.

Thanks,
Prasanga Dhungel

[RenYang-home]

What test sequences are you looking for? JCT-VC dataset?

[prasangadhungel]

Dear Ren,

We are thankful for you help and support. Please refer the experiment for the experimental results. As suspected our PSNR optimized model suffers a drop in performance but MS-SSIM optimized models seem to stand the ground.

Regards,
Prasanga Dhungel

[RenYang-home]

Dear Prasanga,Thanks for informing me your updates. Please note that the MS-SSIM curves in the original DVC paper are the results of the PSNR trained models. DVC did not train MS-SSIM optimized model in their CVPR paper, but did it in their PAMI paper, refer tohttps://ieeexplore.ieee.org/abstract/document/9072487Best,Ren-------- 原始邮件 --------发件人： PrasangaDhungel <notifications@github.com>日期： 2020年10月9日周五 15:40收件人： tukilabs/Video-Compression-Net <Video-Compression-Net@noreply.github.com>抄送： Ren Yang <r.yangchn@gmail.com>, Author <author@noreply.github.com>主 题： Re: [tukilabs/Video-Compression-Net] Motion compensation (#1) Dear Ren, We are thankful for you help and support. Please refer the experiment for the experimental results. As suspected our PSNR optimized model suffers a drop in performance but MS-SSIM optimized models seem to stand the ground. Regards, Prasanga Dhungel —You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.

[prasangadhungel]

Dear Ren,

Thanks for the information, we will look into this further.

Regards,
Prasanga Dhungel