Metraipolis › archive › dc_39376a70ca97

Efficient 3D Gaussian Head Avatars for Edge Devices — arXiv:2610.09821

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09821
License (as archived)
CC0-1.0
Archived
2026-10-08T12:47:39.387Z
SHA-256 of the archived text
7dc77fb9dd1049cae2c63ccb5f447d1cb6f8325519352240f4cf5517ce0fcc5c
Stable URL of this page
https://metraipolis.com/documento/dc_39376a70ca97

How to cite

Efficient 3D Gaussian Head Avatars for Edge Devices — arXiv:2610.09821 — arXiv (en), https://arxiv.org/abs/2610.09821 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_39376a70ca97 (Metraipolis archive dc_39376a70ca97, sha256 7dc77fb9dd1049cae2c63ccb5f447d1cb6f8325519352240f4cf5517ce0fcc5c)
BibTeX
@misc{dc39376a70ca97,
  title = {Efficient 3D Gaussian Head Avatars for Edge Devices — arXiv:2610.09821},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09821},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_39376a70ca97; sha256 7dc77fb9dd1049cae2c63ccb5f447d1cb6f8325519352240f4cf5517ce0fcc5c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09821. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Generative 3D Gaussian head avatars provide high-quality, efficient rendering, but synthesising the Gaussian representation remains computationally expensive, limiting deployment on resource-constrained and edge devices. We introduce an efficient generator architecture for unconditional 3D Gaussian head synthesis, based on a parameter-efficient synthesis block and depth-wise separable convolutions while retaining style-based conditioning. Our architecture reduces generator complexity without requiring model compression or quantisation. Compared with the baseline model, our approach reduces FLOPs by 94%, parameter count by 70%, and model size by 81%, while maintaining competitive generation quality. We further demonstrate practical CPU inference and browser-based execution on mobile devices using ONNX Runtime, enabling 3D Gaussian avatar synthesis without dedicated GPU hardware or application-specific software. In addition to conventional image-quality metrics, we evaluate multi-view consistency, training cost, and deployment performance. Code, trained models, and evaluation tools will be released publicly.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.09821 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T12:18:15.421Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).