Sangbeom Lim, Seoung Wug Oh, Heeji Yoon, Seungryong Kim, Joon-Young Lee · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.26857 · cited by 0 works (OpenAlex) · open-access PDF
Длиннометражные видеоматериалы недавно появились в качестве лидирующего подхода, используя видеоматериалы для диффузии с интернета, чтобы добиться сильной нулевой генерализации из ограниченных синтетических данных. Тем не менее, тяжелый вычисление этих диффузионных спинников ограничивает вывод до короткого куска всего от 6 до 24 кадров, поэтому даже обычные видеоматериалы охватывают несколько кусков. Поскольку каждый кусок матрицы представлен как отдельные токенные потоки, которые взаимодействуют независимо от грубого маскирования, существующие методы не поддерживают согласованность через границы куска, производя финиковые границы и непрерывную прозрачность. Мы подходим к этому с помощью каркасной диффузионной рамки, осно
Machine translation (Meta M2M100); the English original follows.
Diffusion-based video matting has recently emerged as a leading approach, leveraging video diffusion priors from internet-scale pretraining to achieve strong zero-shot generalization from limited synthetic data. However, the heavy compute of these diffusion backbones limits inference to a short chunk of just 6 to 24 frames, so even ordinary videos span multiple chunks. Because each chunk is matted independently from coarse mask guidance, existing methods fail to maintain consistency across chunk boundaries, producing flickering edges and discontinuous transparency. We address this with a mask-guided diffusion-matting framework built on two ideas. The RGB and mask conditioning are represented as separate token streams that interact through full spatio-temporal self-attention, letting the model adaptively combine fine RGB texture with coarse mask layout rather than committing to a fixed channel-wise fusion. On top of this, we propose a novel pair of learnable embeddings, the preserve token and the refine token, that act as a per-frame conditioning interface and enable a sliding-window inference scheme in which already-generated mattes are propagated across chunk boundaries for local temporal consistency, complemented by a globally shared reference anchor at every chunk to prevent error accumulation in long sequences without any additional training. Across standard video matting benchmarks and in-the-wild long footage, our framework outperforms prior diffusion-based methods in both fine-detail accuracy and temporal consistency, and uniquely maintains consistency across chunk boundaries on long videos.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Security token, Computer science, Inference, Generalization, Consistency (knowledge bases), Artificial intelligence
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26857, 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