cleardusk

5 ms·

1. 1.▲
   by cleardusk 4mo ago

   vllm-omni support is on the way : )
2. 2.▲
   by cleardusk 4mo ago

   :D
3. 3.▲
   Show HN: Lance – image/video generation and understanding in one model(github.com)

   64 points by cleardusk 4mo ago|15 comments
4. 4.▲
   by cleardusk 2y ago

   I scanned the open codec page: https://en.wikipedia.org/wiki/List_of_open-source_codecs I'm a little confused, is H.265 not OPEN? :-(
5. 5.▲
   by cleardusk 2y ago

   Thanks for the reminder. The homepage is a GitHub page and does not support git lfs, so I have compressed the files as much as possible to reduce their size. We consider re-encode the mp4 files to x264, and provide a packed zip of the homep
6. 6.▲
   by cleardusk 2y ago

   Understood. The core dependency of InsightFace in LivePortrait is the face detection algo. The face detection is easily to be replaced with self-developed or MIT-licensed model.
7. 7.▲
   by cleardusk 2y ago

   https://github.com/deepinsight/insightface?tab=readme-ov-fil... ---------- The code of InsightFace is released under the MIT License. There is no limitation for both academic and commercial usage.
8. 8.▲
   by cleardusk 2y ago

   Fixed! h_h
9. 9.▲
   LivePortrait: A fast, controllable portrait animation model(github.com)

   203 points by cleardusk 2y ago|25 comments
10. 10.▲
    by cleardusk 6y ago

    :)
11. 11.▲
    by cleardusk 6y ago

    Not a random face. But it may not look very similar.
12. 12.▲
    by cleardusk 6y ago

    The access is OK.
13. 13.▲
    by cleardusk 6y ago

    Thanks for your detailed reply and explaination. I learned from it.
14. 14.▲
    by cleardusk 6y ago

    Thanks, I will consider adding it later :)
15. 15.▲
    by cleardusk 6y ago

    Thanks for your advice. I have no experience on a cooperative project before, so my committed messages are kind of meaningless. I think you are right. I will read the committing guidance and take care of them later. Thanks for your critica
16. 16.▲
    by cleardusk 6y ago

    Thanks for your advice : )
17. 17.▲
    by cleardusk 6y ago

    Thanks for your interest and try. The theory computation complexity is described in the paper, it is rather small. However, whether achieving real-time really depends on your hardware, your need and the code optimization.
18. 18.▲
    by cleardusk 6y ago

    thanks for your interest, releasing the training code should be granted by my lab leader. I am trying to apply for it. If you have any question, welcome to raise an issue or email me :)
19. 19.▲
    by cleardusk 6y ago

    Welcome to give it a try :)
20. 20.▲
    3D face: fast, accurate and stable reconstruction(github.com)

    146 points by cleardusk 6y ago|21 comments
21. 21.▲
    by cleardusk 8y ago

    I am sorry for my fault. I will delete them soon. In other words, the license is valid in all the world, including China.
22. 22.▲
    by cleardusk 8y ago

    Thank you for your warning. I have ignored it before and I will delete the modified BFM data soon.
23. 23.▲
    Real-time 3D Face Reconstruction(github.com)

    126 points by cleardusk 8y ago|11 comments
24. 24.▲
    Face Recognition: an eyeglass face dataset(github.com)

    3 points by cleardusk 8y ago|0 comments
25. 25.▲
    Super Real-Time 3D Dense Face Alignment in Full Pose Range(github.com)

    3 points by cleardusk 8y ago|0 comments