[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-2116":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":15,"stars7d":16,"stars30d":17,"stars90d":14,"forks30d":14,"starsTrendScore":18,"compositeScore":19,"rankGlobal":9,"rankLanguage":9,"license":20,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":23,"hasPages":21,"topics":24,"createdAt":9,"pushedAt":9,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":14,"starSnapshotCount":14,"syncStatus":28,"lastSyncTime":29,"discoverSource":30},2116,"TDMM-LM_data","Songluchuan\u002FTDMM-LM_data","Songluchuan","[CVPR2026] TDMM-LM: Bridging Facial Understanding and Animation via Language Models",null,"Python",489,9,8,0,39,91,268,117,93,"MIT License",false,"main",true,[],"2026-06-12 04:00:13","\n# ✨ TDMM-LM: Bridging Facial Understanding and Animation via Language Models\n[**🌐 Homepage**](https:\u002F\u002Fsongluchuan.github.io\u002FTDMM-LM\u002F) | [**🔬 Paper**](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.16936) | [**👩‍💻 Code**](https:\u002F\u002Fgithub.com\u002FSongluchuan\u002FTDMM-LM_data)\n\n## TDMM-LM Dataset\n> TDMM-LM Dataset is a large-scale facial animation dataset synthesized with foundation generative models, comprising roughly 80 hours of face-centric video that spans a wide spectrum of emotions,  expressions, and head motions, with each clip paired with its text prompt and 3D facial parameters for training text-driven facial animation\u002Funderstanding models.\n\n![alt text](assets\u002FTDMMLM.png)\n\n\nOur dataset enables researchers and practitioners to uncover the strengths, limitations, and potential areas for improvement in text-driven facial animation\u002Funderstaning models, offering valuable insights into the challenges of generating expressive and emotionally faithful facial behavior.\n\n\n\n## 📊 Video Dataset\u002FAnnotation [Part-1, \\~70hr]\n\n\n• Videos Download: [**Google drive**](https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Ffolders\u002F11wWL6vWxxzHJMpSzYlA0uijJkCO10mLC?usp=sharing) (.\u002Fdownload_gdrive_folder.sh)\n\n• Language Annotation: As shown in [**json file**](https:\u002F\u002Fgithub.com\u002FSongluchuan\u002FTDMM-LM_data\u002Fblob\u002Fmain\u002Fjson\u002Fdata_part1.json).\n\n## 📊 Video Dataset\u002FAnnotation [Part-2, \\~10hr]\n\n\n• Coming Soon.\n\n## 🎵 Audios\n\n• Coming Soon [Synchronized with videos in Part-1].\n\n## 🔧 Tools\n\n• We recommend using [**smirk**](https:\u002F\u002Fgithub.com\u002Fgeorgeretsi\u002Fsmirk) or other facial tracking methods to extract the parameters. \n\n• We provide a [**batch processing script by smirk**](https:\u002F\u002Fgithub.com\u002FSongluchuan\u002FTDMM-LM_data\u002Ftree\u002Fmain\u002Ftools\u002Fsmirk_inverse) as a reference. \n\n• We provide a [**batch processing script by spectre**](https:\u002F\u002Fgithub.com\u002FSongluchuan\u002FTDMM-LM_data\u002Ftree\u002Fmain\u002Ftools\u002Fspectre_inverse) as a reference. \n\n\n## ✏️ Citation\n```bibtex\n@article{song2026tdmm,\n  title={TDMM-LM: Bridging Facial Understanding and Animation via Language Models},\n  author={Song, Luchuan and Liu, Pinxin and Liu, Haiyang and Jin, Zhenchao and Tang, Yolo Yunlong and Xu, Zichong and Liang, Susan and Bi, Jing and Corso, Jason J and Xu, Chenliang},\n  journal={arXiv preprint arXiv:2603.16936},\n  year={2026}\n}\n```\n","TDMM-LM是一个通过语言模型连接面部理解和动画的大规模数据集。该项目提供了约80小时的面部中心视频，覆盖了广泛的情感、表情和头部动作，并为每个视频片段配以文本提示和3D面部参数，用于训练文本驱动的面部动画或理解模型。其核心功能包括高质量的合成数据生成与详细的注释支持，采用Python开发并开放MIT许可证。适用于研究者和从业者探索文本驱动下的人脸动画技术潜力及其局限性，特别是在提高面部表情的真实性和情感准确性方面具有重要价值。",2,"2026-06-11 02:48:14","CREATED_QUERY"]