Metraipolis › archive › dc_04497639fdfe
UniRGBA: Unifying RGB and Alpha in Pretrained Video Diffusion Models — arXiv:2611.09069
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.09069
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-09T22:34:48.132Z
- SHA-256 of the archived text
2a146c8f28d3b48df9a79bd35d30736145e5a4e67714c364d881767fc8acd39e- Stable URL of this page
https://metraipolis.com/documento/dc_04497639fdfe
How to cite
UniRGBA: Unifying RGB and Alpha in Pretrained Video Diffusion Models — arXiv:2611.09069 — arXiv (en), https://arxiv.org/abs/2611.09069 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_04497639fdfe (Metraipolis archive dc_04497639fdfe, sha256 2a146c8f28d3b48df9a79bd35d30736145e5a4e67714c364d881767fc8acd39e)
BibTeX
@misc{dc04497639fdfe,
title = {UniRGBA: Unifying RGB and Alpha in Pretrained Video Diffusion Models — arXiv:2611.09069},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.09069},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_04497639fdfe; sha256 2a146c8f28d3b48df9a79bd35d30736145e5a4e67714c364d881767fc8acd39e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09069. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Recent advances in video diffusion models have achieved remarkable photorealistic RGB synthesis, yet generating videos with a native alpha channel remains an open challenge. Existing approaches often rely on matting-style data and non-native RGBA designs, causing them to specialize in real-object scenes and to degrade severely on visual-effect content whose semi-transparency is spatially non-local and non-elliptical. We present UniRGBA, a unified framework whose key insight is that RGBA generation does not require a dedicated alpha encoder or complex latent-space restructuring---a minimal, targeted extension of the existing RGB pathway is sufficient. UniRGBA-VAE extends a causal video VAE to 4-channel RGBA via zero-initialized channel expansion, symmetric selective layer unfreezing (5% trainable), and a six-term progressive reconstruction loss anchored by a KL reference loss that prevents latent drift, followed by a two-stage schedule that introduces a lightweight GAN objective. We then apply standard DoRA to the pretrained DiT, reusing the unified 4-channel latent without any alpha-specific architectural changes. On a visual-effect RGBA benchmark, UniRGBA substantially outperforms Wan-Alpha on RGBA reconstruction while preserving RGB fidelity. For generation, UniRGBA outperforms Wan-Alpha on RGBA text-to-video tasks and matches or surpasses the original pretrained Wan2.1 backbone on RGB-only metrics, demonstrating that native alpha support can be integrated without compromising the underlying RGB prior or the model's architectural versatility. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.09069 — 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 2026-11-09T22:30:47.317Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).