Yuanxiang Ni, Xianliang Huang, Ruihan Hou, Feng Xie, Hao Zhang · 2026 · arXiv
DOI: 10.48550/arXiv.2610.28351 · open-access PDF
Rekonstruksi adegan 3D dinamis dari video monokular membutuhkan pengelompokan gerakan dari pengamatan yang berisik dan sering terlihat. track jangka panjang memberikan keterbatasan geometris yang berharga, tetapi mengintegrasikan cakupan waktu yang terfragmentasi ke dalam gerakan berkelanjutan bersama tetap menantang. Kami memperkenalkan \textbf{RayCo4D}, model gerakan berkelanjutan yang dikondisikan oleh sinar untuk rekonstruksi 4D monokular yang secara eksplisit memisahkan pengukuran visual dari keadaan gerakan 3D yang tersembunyi.
Machine translation (Meta M2M100); the English original follows.
Reconstructing dynamic 3D scenes from monocular video requires inferring motion from noisy and intermittently visible observations. Long-range tracks provide valuable geometric constraints, but integrating their fragmented temporal coverage into shared, continuous motion remains challenging. We introduce \textbf{RayCo4D}, a ray-conditioned continuous motion model for monocular 4D reconstruction that explicitly separates visual measurements from latent 3D motion states. Each track is associated with a reference-time 3D query constrained to its camera ray, and its image and depth observations co
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=2610.28351, 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