https://github.com/foundationvision/omnitokenizer

[NeurIPS 2024]OmniTokenizer: one model and one weight for image-video joint tokenization.
https://github.com/foundationvision/omnitokenizer

auto-regressive-model image-generation tokenization vae video-generation vqvae

Last synced: about 1 month ago
JSON representation

[NeurIPS 2024]OmniTokenizer: one model and one weight for image-video joint tokenization.

• Host: GitHub
• URL: https://github.com/foundationvision/omnitokenizer
• Owner: FoundationVision
• License: mit
• Created: 2024-06-13T15:08:47.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2024-07-09T17:00:57.000Z (11 months ago)
• Last Synced: 2025-03-31T14:09:04.416Z (about 2 months ago)
• Topics: auto-regressive-model, image-generation, tokenization, vae, video-generation, vqvae
• Language: Python
• Homepage: https://www.wangjunke.info/OmniTokenizer/
• Size: 68.9 MB
• Stars: 286
• Watchers: 5
• Forks: 7
• Open Issues: 8
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project