Newest papers

Stream4D: Streaming 4D Reconstruction with Training-free Motion Decoupling via Causal KV-Cache Attention

Wei Zhang, Ying Feng, Mengtong Wu, Shiqiang Gong, Huaqi Zhang, Qi Zhang · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.31320 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Rekonstruksi adegan 4D membutuhkan pengolahan objek dinamis yang kuat, namun metode streaming yang ada biasanya mengasumsikan adegan statis, yang mengarah ke pos drift dan artifak ghosting selama urutan panjang. Kami memperkenalkan Stream4D, framework rekonstruksi streaming bebas pelatihan, yang dapat diintegrasikan secara langsung ke tulang belakang Transformer penyebab (misalnya, StreamVGGT, Stream3R). Pengamatan utama kami adalah bahwa perhatian penyebab secara implisit mengkodekan tanda-tanda gerak sementara yang kaya di dalam cache KV. Kami secara adaptif mengumpulkan tanda-tanda ini dari berbagai lapisan perhatian dan di seluruh jendela waktu, dan menggabungkan mereka dengan priors mendalam untuk menangani kasus yang menantang seperti oklusion dan wilayah tanpa tekstur, menghasilkan masks dinamis yang akurat yang memecahkan wilayah statik dan dinamis. Kami kemudian menerapkan

Machine translation (Meta M2M100); the English original follows.

Abstract

4D scene reconstruction requires robust handling of dynamic objects, yet existing streaming methods commonly assume static scenes, leading to pose drift and ghosting artifacts over long sequences. We introduce Stream4D, a training-free, motion-decoupled streaming reconstruction framework that can be directly integrated into causal Transformer backbones (e.g., StreamVGGT, Stream3R). Our key observation is that causal attention implicitly encodes rich temporal motion cues within the KV cache. We adaptively aggregate these cues from multiple attention layers and across temporal windows, and fuse them with depth priors to handle challenging cases such as occlusions and textureless regions, producing precise dynamic masks that decouple static and dynamic regions. We then apply a soft key masking scheme that leverages these dynamic masks to suppress interference from dynamic regions in causal attention, thereby improving camera pose estimation accuracy. Stream4D inherits the efficiency of causal attention and remains compatible with existing streaming architectures. On diverse dynamic-scene benchmarks, Stream4D significantly improves camera pose estimation and static-scene reconstruction quality, yields reliable dynamic masks, and effectively mitigates long-sequence ghosting.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Ghosting, Computer vision, Artificial intelligence, Key (lock), Fuse (electrical)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange