Mohit Mendiratta

I lead the AI research team at QOVES, where we build facial analysis and AI-powered image editing products. My team of ML researchers and engineers works on taking 3D face models, generative editing and vision-language models from research into production.

My research sits at the intersection of computer vision and computer graphics, with a focus on photorealistic 3D faces and heads. I work on morphable models and Gaussian splatting head avatars, text-driven editing of dynamic 3D humans, and vision-language models that can reason about faces. Recent work includes GRMM, a real-time Gaussian morphable head model with learned residuals (3DV 2026), and AvatarStudio, text-driven editing of dynamic head avatars (SIGGRAPH Asia 2023).

I did my PhD research at the Max Planck Institute for Informatics in the Visual Computing and Artificial Intelligence department, advised by Christian Theobalt, from 2021 to 2026; the Dr.-Ing. defence is pending. Before that I completed an MSc in Visual Computing at Saarland University and a BTech in Electrical and Electronics Engineering at KIIT, Bhubaneswar. I review for SIGGRAPH, SIGGRAPH Asia, CVPR, 3DV and Eurographics.

Email Google Scholar GitHub LinkedIn CV (PDF)

selected publications

ECCV

EmoteGPT: 3D Human Facial Expressions from Natural Language Descriptions

Haoran Wang, Mohit Mendiratta, Christian Theobalt, and Adam Kortylewski

In European Conference on Computer Vision (ECCV), 2026

arXiv Bib HTML Video Code

@inproceedings{wang2026emotegpt,
  title = {EmoteGPT: 3D Human Facial Expressions from Natural Language Descriptions},
  author = {Wang, Haoran and Mendiratta, Mohit and Theobalt, Christian and Kortylewski, Adam},
  booktitle = {European Conference on Computer Vision (ECCV)},
  year = {2026},
}

3DV

GRMM: Real-Time High-Fidelity Gaussian Morphable Head Model with Learned Residuals

Mohit Mendiratta, Mayur Deshmukh, Kartik Teotia, Vladislav Golyanik, Adam Kortylewski, and Christian Theobalt

In International Conference on 3D Vision (3DV), 2026

arXiv Bib HTML

@inproceedings{mendiratta2026grmm,
  title = {GRMM: Real-Time High-Fidelity Gaussian Morphable Head Model with Learned Residuals},
  author = {Mendiratta, Mohit and Deshmukh, Mayur and Teotia, Kartik and Golyanik, Vladislav and Kortylewski, Adam and Theobalt, Christian},
  booktitle = {International Conference on 3D Vision (3DV)},
  year = {2026},
}

SIGGRAPH Asia

Audio-Driven Universal Gaussian Head Avatars

Kartik Teotia, Helge Rhodin, Mohit Mendiratta, Hyeongwoo Kim, Marc Habermann, and Christian Theobalt

In SIGGRAPH Asia 2025 Conference Papers, 2025

arXiv Bib HTML

@inproceedings{teotia2025unigaha,
  title = {Audio-Driven Universal Gaussian Head Avatars},
  author = {Teotia, Kartik and Rhodin, Helge and Mendiratta, Mohit and Kim, Hyeongwoo and Habermann, Marc and Theobalt, Christian},
  booktitle = {SIGGRAPH Asia 2025 Conference Papers},
  year = {2025},
}

CVPR

Zero-Shot Video Semantic Segmentation based on Pre-Trained Diffusion Models

Qian Wang, Abdelrahman Eldesokey, Mohit Mendiratta, Fangneng Zhan, Adam Kortylewski, Christian Theobalt, and 1 more author

In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025

arXiv Bib HTML

@inproceedings{wang2025zeroshotvss,
  title = {Zero-Shot Video Semantic Segmentation based on Pre-Trained Diffusion Models},
  author = {Wang, Qian and Eldesokey, Abdelrahman and Mendiratta, Mohit and Zhan, Fangneng and Kortylewski, Adam and Theobalt, Christian and Wonka, Peter},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2025},
}

3DV

TEDRA: Text-Based Editing of Dynamic and Photoreal Actors

Basavaraj Sunagad, Heming Zhu^*, Mohit Mendiratta^*, Adam Kortylewski, Christian Theobalt, and Marc Habermann

In International Conference on 3D Vision (3DV), 2025

arXiv Bib HTML

@inproceedings{sunagad2025tedra,
  title = {TEDRA: Text-Based Editing of Dynamic and Photoreal Actors},
  author = {Sunagad, Basavaraj and Zhu, Heming and Mendiratta, Mohit and Kortylewski, Adam and Theobalt, Christian and Habermann, Marc},
  booktitle = {International Conference on 3D Vision (3DV)},
  pages = {1451--1461},
  year = {2025},
}

SIGGRAPH Asia

AvatarStudio: Text-driven Editing of 3D Dynamic Human Head Avatars

Mohit Mendiratta, Xingang Pan^*, Mohamed Elgharib^*, Kartik Teotia, Mallikarjun B R, Ayush Tewari, and 3 more authors

ACM Transactions on Graphics (Proc. SIGGRAPH Asia), 2023

DOI arXiv Bib HTML Video

@article{mendiratta2023avatarstudio,
  title = {AvatarStudio: Text-driven Editing of 3D Dynamic Human Head Avatars},
  author = {Mendiratta, Mohit and Pan, Xingang and Elgharib, Mohamed and Teotia, Kartik and B R, Mallikarjun and Tewari, Ayush and Golyanik, Vladislav and Kortylewski, Adam and Theobalt, Christian},
  journal = {ACM Transactions on Graphics (Proc. SIGGRAPH Asia)},
  volume = {42},
  number = {6},
  articleno = {226},
  year = {2023},
  doi = {10.1145/3618368},
}

SIGGRAPH Asia

Egocentric Videoconferencing

Mohamed Elgharib^*, Mohit Mendiratta^*, Justus Thies, Matthias Nießner, Hans-Peter Seidel, Ayush Tewari, and 2 more authors

ACM Transactions on Graphics (Proc. SIGGRAPH Asia), 2020

arXiv Bib HTML PDF Supp Video

@article{elgharib2020egochat,
  title = {Egocentric Videoconferencing},
  author = {Elgharib, Mohamed and Mendiratta, Mohit and Thies, Justus and Nie{\ss}ner, Matthias and Seidel, Hans-Peter and Tewari, Ayush and Golyanik, Vladislav and Theobalt, Christian},
  journal = {ACM Transactions on Graphics (Proc. SIGGRAPH Asia)},
  volume = {39},
  number = {6},
  articleno = {268},
  year = {2020},
}

in the media

Two Minute Papers on Egocentric Videoconferencing (SIGGRAPH Asia 2020)

YouTube Short on Beauty is in the AI of the beholder (arXiv 2026)