Shuai Zhang, Zhanpeng Ouyang, Zhengqing Chen, Ruohong Mei, Shaoyu Chen, Weiqiang Ren, Hui Sun, Qian Zhang · 2026 · arXiv
DOI: 10.48550/arXiv.2610.28521 · open-access PDF
A geração de vídeo e a reconstrução 3D de feed-forward normalmente usam representações separadas, exigindo decodificação RGB e re-codificação visual para recuperar a geometria dos latentes gerados. Proponemos a Ponte Latente, que conecta espaços latentes de geração de vídeo com espaços de características geométricas para prever diretamente a geometria 3D dos latentes de vídeo. Desenvolvemos pontes específicas de representação para Wan2.1 e V-RAE, permitindo a previsão de feed-forward de câmeras, profundidade e nuvens de ponto 3D sem a via RGB intermediária.
Machine translation (Meta M2M100); the English original follows.
Video generation and feed-forward 3D reconstruction typically use separate representations, requiring RGB decoding and visual re-encoding to recover geometry from generated latents. We propose Latent Bridge, which connects video generative latent spaces with geometric feature spaces to directly predict 3D geometry from video latents. We develop representation-specific bridges for Wan2.1 and V-RAE, enabling feed-forward prediction of cameras, depth, and 3D point clouds without the intermediate RGB pathway. Each bridge is trained on RGB videos through feature distillation, while the video encode
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.28521, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange