Open eyaler opened 9 months ago
could you share the RVM pipeline used in the paper?
Could you please explain what 'RVM pipeline' means in detail?
I am trying to build the pipeline for custom videos.
"We also evaluate the proposed method on some self-collected datasets, utilizing the video matting method [30] to generate the foreground videos. "
[30] Shanchuan Lin, Linjie Yang, Imran Saleemi, and Soumyadip Sengupta. Robust high-resolution video matting with temporal guidance. CoRR, 2021
I am trying to build the pipeline for custom videos.
"We also evaluate the proposed method on some self-collected datasets, utilizing the video matting method [30] to generate the foreground videos. "
[30] Shanchuan Lin, Linjie Yang, Imran Saleemi, and Soumyadip Sengupta. Robust high-resolution video matting with temporal guidance. CoRR, 2021
Hello, since the training of NLA requires a video mask, the custom video I create is simply a segmented video using the video matting method to obtain the video mask, which can be accomplished by following https://github.com/PeterL1n/RobustVideoMatting. Alternatively, you can use https://github.com/gaomingqi/Track-Anything to segment your own video.
could you share the RVM pipeline used in the paper?