解决长序列3D重建中精度衰减问题，实现千米级在线重建
LongStream: Long-Sequence Streaming Autoregressive Visual Geometry
- 不依赖首帧定位，改用关键帧相对位姿预测
- 分离尺度与几何估计，抑制尺度漂移，支持千米级重建
- 优化注意力机制，提升超长序列推理稳定性
长序列流式3D重建仍是重大挑战。现有自回归模型在处理长序列时因锚定首帧位姿，导致注意力衰减、尺度漂移和外推误差。我们提出LongStream，一种新型无参照系流式视觉几何模型，可在严格在线、未来不可见条件下实现数千帧的度量尺度场景重建。方法上：第一，放弃首帧锚定，预测关键帧相对位姿，将长程外推转化为恒定难度的局部任务；第二，引入正交尺度学习，完全解耦几何与尺度估计，抑制漂移；第三，识别Transformer中的注意力偏差问题，包括注意力汇聚依赖和长期键值缓存饱和，提出缓存一致性训练与周期性缓存刷新策略。该方法有效抑制超长序列中的注意力偏差与污染，缩小训练与推理差距。实验表明，LongStream达到当前最优性能，在18 FPS下实现千米级序列稳定度量重建。
原文摘要 · Abstract (English)
Long-sequence streaming 3D reconstruction remains a significant open challenge. Existing autoregressive models often fail when processing long sequences because they anchor poses to the first frame, leading to attention decay, scale drift, and extrapolation errors. We introduce LongStream, a novel gauge-decoupled streaming visual geometry model for metric-scale scene reconstruction across thousands of frames under a strictly online, future-invisible setting. Our approach is threefold. First, we discard the first-frame anchor and predict keyframe-relative poses. This reformulates long-range extrapolation into a constant-difficulty local task. Second, we introduce orthogonal scale learning. This method fully disentangles geometry from scale estimation to suppress drift. Finally, we identify attention bias issues in Transformers, including attention-sink reliance and long-term KV-cache saturation. We propose cache-consistent training combined with periodic cache refresh. This approach suppresses attention biases and contamination over ultra-long sequences and reduces the gap between training and inference. Experiments show that LongStream achieves state-of-the-art performance, enabling stable, metric-scale reconstruction over kilometer-scale sequences at 18 FPS. Project Page: https://3dagentworld.github.io/longstream/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。