https://whwjdqls.github.io/discord.github.io/\n","updatedAt":"2024-12-02T09:40:41.199Z","author":{"_id":"64c2c45ae818eec6128fdda3","avatarUrl":"/avatars/d4399e25e6399345e263c7902789047e.svg","fullname":"Junwan Kim","name":"junwann","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":4}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.9093340635299683},"editors":["junwann"],"editorAvatarUrls":["/avatars/d4399e25e6399345e263c7902789047e.svg"],"reactions":[],"isReport":false}},{"id":"674e6033288854089112372c","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":264},"createdAt":"2024-12-03T01:34:43.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"This is an automated message from the [Librarian Bot](https://huggingface.co/librarian-bots). I found the following papers similar to this paper. \n\nThe following papers were recommended by the Semantic Scholar API \n\n* [Rethinking Diffusion for Text-Driven Human Motion Generation](https://huggingface.co/papers/2411.16575) (2024)\n* [MotionGPT-2: A General-Purpose Motion-Language Model for Motion Generation and Understanding](https://huggingface.co/papers/2410.21747) (2024)\n* [MotionLLaMA: A Unified Framework for Motion Synthesis and Comprehension](https://huggingface.co/papers/2411.17335) (2024)\n* [LaMP: Language-Motion Pretraining for Motion Generation, Retrieval, and Captioning](https://huggingface.co/papers/2410.07093) (2024)\n* [Hallo2: Long-Duration and High-Resolution Audio-Driven Portrait Image Animation](https://huggingface.co/papers/2410.07718) (2024)\n* [LEAD: Latent Realignment for Human Motion Diffusion](https://huggingface.co/papers/2410.14508) (2024)\n* [Enhancing Motion in Text-to-Video Generation with Decomposed Encoding and Conditioning](https://huggingface.co/papers/2410.24219) (2024)\n\n\n Please give a thumbs up to this comment if you found it helpful!\n\n If you want recommendations for any Paper on Hugging Face checkout [this](https://huggingface.co/spaces/librarian-bots/recommend_similar_papers) Space\n\n You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: `@librarian-bot recommend`","html":"
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
\n
The following papers were recommended by the Semantic Scholar API
Please give a thumbs up to this comment if you found it helpful!
\n
If you want recommendations for any Paper on Hugging Face checkout this Space
\n
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend
\n","updatedAt":"2024-12-03T01:34:43.172Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":264}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6960901021957397},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2411.19527","authors":[{"_id":"674d17baaa794726fac6a6c9","user":{"_id":"65b5d43419d2172f003fed47","avatarUrl":"/avatars/3683356c1da06ddfe93fa674aa4481f0.svg","isPro":false,"fullname":"jbc","user":"whwjdqls99","type":"user"},"name":"Jungbin Cho","status":"claimed_verified","statusLastChangedAt":"2024-12-02T08:49:56.163Z","hidden":false},{"_id":"674d17baaa794726fac6a6ca","user":{"_id":"64c2c45ae818eec6128fdda3","avatarUrl":"/avatars/d4399e25e6399345e263c7902789047e.svg","isPro":false,"fullname":"Junwan Kim","user":"junwann","type":"user"},"name":"Junwan Kim","status":"claimed_verified","statusLastChangedAt":"2024-12-02T08:49:53.178Z","hidden":false},{"_id":"674d17baaa794726fac6a6cb","name":"Jisoo Kim","hidden":false},{"_id":"674d17baaa794726fac6a6cc","user":{"_id":"64f22ee8a2fc808156dd1872","avatarUrl":"/avatars/45d9140acbfdbb0a8b044c2f9d3c4fad.svg","isPro":false,"fullname":"Kim Minseo","user":"min99830","type":"user"},"name":"Minseo Kim","status":"claimed_verified","statusLastChangedAt":"2024-12-02T12:52:35.160Z","hidden":false},{"_id":"674d17baaa794726fac6a6cd","user":{"_id":"66388b9b7517eefed69a13da","avatarUrl":"/avatars/bbac095ab6e6c1df78b9a9d630a6c207.svg","isPro":false,"fullname":"kangmingu","user":"rkdrn79","type":"user"},"name":"Mingu Kang","status":"claimed_verified","statusLastChangedAt":"2024-12-03T14:40:42.003Z","hidden":false},{"_id":"674d17baaa794726fac6a6ce","name":"Sungeun Hong","hidden":false},{"_id":"674d17baaa794726fac6a6cf","user":{"_id":"674622d01310ed05c6c5a5aa","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/0ga3pvGwd8oTEEWxIuPr6.png","isPro":false,"fullname":"Tae-Hyun Oh","user":"taehyunoh","type":"user"},"name":"Tae-Hyun Oh","status":"claimed_verified","statusLastChangedAt":"2024-12-02T17:31:58.388Z","hidden":false},{"_id":"674d17baaa794726fac6a6d0","name":"Youngjae Yu","hidden":false}],"publishedAt":"2024-11-29T07:54:56.000Z","submittedOnDailyAt":"2024-12-02T07:08:21.019Z","title":"DisCoRD: Discrete Tokens to Continuous Motion via Rectified Flow\n Decoding","submittedOnDailyBy":{"_id":"64c2c45ae818eec6128fdda3","avatarUrl":"/avatars/d4399e25e6399345e263c7902789047e.svg","isPro":false,"fullname":"Junwan Kim","user":"junwann","type":"user"},"summary":"Human motion, inherently continuous and dynamic, presents significant\nchallenges for generative models. Despite their dominance, discrete\nquantization methods, such as VQ-VAEs, suffer from inherent limitations,\nincluding restricted expressiveness and frame-wise noise artifacts. Continuous\napproaches, while producing smoother and more natural motions, often falter due\nto high-dimensional complexity and limited training data. To resolve this\n\"discord\" between discrete and continuous representations, we introduce\nDisCoRD: Discrete Tokens to Continuous Motion via Rectified Flow Decoding, a\nnovel method that decodes discrete motion tokens into continuous motion through\nrectified flow. By employing an iterative refinement process in the continuous\nspace, DisCoRD captures fine-grained dynamics and ensures smoother and more\nnatural motions. Compatible with any discrete-based framework, our method\nenhances naturalness without compromising faithfulness to the conditioning\nsignals. Extensive evaluations demonstrate that DisCoRD achieves\nstate-of-the-art performance, with FID of 0.032 on HumanML3D and 0.169 on\nKIT-ML. These results solidify DisCoRD as a robust solution for bridging the\ndivide between discrete efficiency and continuous realism. Our project page is\navailable at: https://whwjdqls.github.io/discord.github.io/.","upvotes":11,"discussionId":"674d17bcaa794726fac6a755","projectPage":"https://whwjdqls.github.io/discord.github.io/","githubRepo":"https://github.com/whwjdqls/DisCoRD","ai_summary":"DisCoRD enhances human motion generation by translating discrete motion tokens into continuous motion through rectified flow, achieving state-of-the-art performance.","ai_keywords":["VQ-VAEs","DisCoRD","rectified flow","continuous motion","iterative refinement","HumanML3D","KIT-ML","FID"],"githubStars":38},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64c2c45ae818eec6128fdda3","avatarUrl":"/avatars/d4399e25e6399345e263c7902789047e.svg","isPro":false,"fullname":"Junwan Kim","user":"junwann","type":"user"},{"_id":"64cb5884d469fc2cf83bdd76","avatarUrl":"/avatars/10e63cf62d8200beef3e31846796e398.svg","isPro":false,"fullname":"JisooKim","user":"Jiiiiiisoo","type":"user"},{"_id":"63be1cd13b0665ad51d29c37","avatarUrl":"/avatars/5acc9b9bbecac3d567e927e2d8667b00.svg","isPro":false,"fullname":"Seungwon Lim","user":"sngwon","type":"user"},{"_id":"648eb1eb59c4e5c87dc116e0","avatarUrl":"/avatars/c636cea39c2c0937f01398c94ead5dad.svg","isPro":false,"fullname":"fdsqefsgergd","user":"T-representer","type":"user"},{"_id":"64f22ee8a2fc808156dd1872","avatarUrl":"/avatars/45d9140acbfdbb0a8b044c2f9d3c4fad.svg","isPro":false,"fullname":"Kim Minseo","user":"min99830","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"64d98ef7a4839890b25eb78b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64d98ef7a4839890b25eb78b/215-CSVLl81z6CAq0ECWU.jpeg","isPro":true,"fullname":"Fangyuan Yu","user":"Ksgk-fy","type":"user"},{"_id":"66388b9b7517eefed69a13da","avatarUrl":"/avatars/bbac095ab6e6c1df78b9a9d630a6c207.svg","isPro":false,"fullname":"kangmingu","user":"rkdrn79","type":"user"},{"_id":"61a5e575dc7a70d5a225be8e","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1679301626301-61a5e575dc7a70d5a225be8e.jpeg","isPro":false,"fullname":"huseyin","user":"hsyntemiz","type":"user"},{"_id":"65c071e569429d85dc5e7e9c","avatarUrl":"/avatars/62a33b17db44e725da4df47ae3d8d554.svg","isPro":false,"fullname":"Jang Han Yoon","user":"jeffrobot","type":"user"},{"_id":"666d506fc0f3d5afc24dd5ca","avatarUrl":"/avatars/eeb98947415d08a26815fd139c76a071.svg","isPro":false,"fullname":"Hyun Ryu","user":"hyun1905","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">
DisCoRD enhances human motion generation by translating discrete motion tokens into continuous motion through rectified flow, achieving state-of-the-art performance.
AI-generated summary
Human motion, inherently continuous and dynamic, presents significant
challenges for generative models. Despite their dominance, discrete
quantization methods, such as VQ-VAEs, suffer from inherent limitations,
including restricted expressiveness and frame-wise noise artifacts. Continuous
approaches, while producing smoother and more natural motions, often falter due
to high-dimensional complexity and limited training data. To resolve this
"discord" between discrete and continuous representations, we introduce
DisCoRD: Discrete Tokens to Continuous Motion via Rectified Flow Decoding, a
novel method that decodes discrete motion tokens into continuous motion through
rectified flow. By employing an iterative refinement process in the continuous
space, DisCoRD captures fine-grained dynamics and ensures smoother and more
natural motions. Compatible with any discrete-based framework, our method
enhances naturalness without compromising faithfulness to the conditioning
signals. Extensive evaluations demonstrate that DisCoRD achieves
state-of-the-art performance, with FID of 0.032 on HumanML3D and 0.169 on
KIT-ML. These results solidify DisCoRD as a robust solution for bridging the
divide between discrete efficiency and continuous realism. Our project page is
available at: https://whwjdqls.github.io/discord.github.io/.
🚀 We are excited to present “DisCoRD: Discrete Tokens to Continuous Motion via Rectified Flow Decoding,” a novel approach to human motion generation that combines the strengths of both discrete and continuous methods.
🏃🏻♂️ Our approach achieves state-of-the-art performance on the HumanML3D and also introduces a novel metric for evaluating the naturalness of generated motion. Feel free to explore our Project page: https://whwjdqls.github.io/discord.github.io/