Autoregressive Appearance Prediction for 3D Gaussian Avatars
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918423711186944 |
|---|---|
| author | Steiner, Michael Chen, Zhang Richard, Alexander Agrawal, Vasu Steinberger, Markus Zollhöfer, Michael |
| author_facet | Steiner, Michael Chen, Zhang Richard, Alexander Agrawal, Vasu Steinberger, Markus Zollhöfer, Michael |
| contents | A photorealistic and immersive human avatar experience demands capturing fine, person-specific details such as cloth and hair dynamics, subtle facial expressions, and characteristic motion patterns. Achieving this requires large, high-quality datasets, which often introduce ambiguities and spurious correlations when very similar poses correspond to different appearances. Models that fit these details during training can overfit and produce unstable, abrupt appearance changes for novel poses. We propose a 3D Gaussian Splatting avatar model with a spatial MLP backbone that is conditioned on both pose and an appearance latent. The latent is learned during training by an encoder, yielding a compact representation that improves reconstruction quality and helps disambiguate pose-driven renderings. At driving time, our predictor autoregressively infers the latent, producing temporally smooth appearance evolution and improved stability. Overall, our method delivers a robust and practical path to high-fidelity, stable avatar driving. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_00928 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Autoregressive Appearance Prediction for 3D Gaussian Avatars Steiner, Michael Chen, Zhang Richard, Alexander Agrawal, Vasu Steinberger, Markus Zollhöfer, Michael Computer Vision and Pattern Recognition Graphics A photorealistic and immersive human avatar experience demands capturing fine, person-specific details such as cloth and hair dynamics, subtle facial expressions, and characteristic motion patterns. Achieving this requires large, high-quality datasets, which often introduce ambiguities and spurious correlations when very similar poses correspond to different appearances. Models that fit these details during training can overfit and produce unstable, abrupt appearance changes for novel poses. We propose a 3D Gaussian Splatting avatar model with a spatial MLP backbone that is conditioned on both pose and an appearance latent. The latent is learned during training by an encoder, yielding a compact representation that improves reconstruction quality and helps disambiguate pose-driven renderings. At driving time, our predictor autoregressively infers the latent, producing temporally smooth appearance evolution and improved stability. Overall, our method delivers a robust and practical path to high-fidelity, stable avatar driving. |
| title | Autoregressive Appearance Prediction for 3D Gaussian Avatars |
| topic | Computer Vision and Pattern Recognition Graphics |
| url | https://arxiv.org/abs/2604.00928 |