LongStream: Long-Sequence Streaming Autoregressive Visual Geometry

• URL: http://arxiv.org/abs/2602.13172v1
• Date: Fri, 13 Feb 2026 18:30:51 GMT
• Title: LongStream: Long-Sequence Streaming Autoregressive Visual Geometry
• Abstract summary: LongStream is a novel gauge-decoupled streaming visual geometry model for metric-scale scene reconstruction.<n>It reformulates long-range extrapolation into a constant-difficulty local task.<n>It delivers stable, metric-scale reconstruction over kilometer-scale sequences at 18 FPS.
• Score: 18.46781332515933
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Long-sequence streaming 3D reconstruction remains a significant open challenge. Existing autoregressive models often fail when processing long sequences. They typically anchor poses to the first frame, which leads 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. 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 solve Transformer cache issues such as attention-sink reliance and long-term KV-cache contamination. We propose cache-consistent training combined with periodic cache refresh. This approach suppresses attention degradation over ultra-long sequences and reduces the gap between training and inference. Experiments show LongStream achieves state-of-the-art performance. It delivers stable, metric-scale reconstruction over kilometer-scale sequences at 18 FPS. Project Page: https://3dagentworld.github.io/longstream/

Related papers

• Revisiting Local Context for Long-Horizon Streaming 3D Reconstruction [10.383154504465232]
  ABot-Recon is a simple streaming model that caches KV features from only the preceding 11 frames.<n>It predicts a point map in the current camera coordinate system together with an adjacent-frame relative pose.<n>To reduce accumulated drift, a lightweight temporal refiner improves relative rotations using recent visual and motion context.
  arXiv  Detail & Related papers  (2026-08-27T15:07:38Z)
• StreamSplat: Streaming Feed-Forward 3D Gaussian Splatting [13.605490241534797]
  StreamSplat is a feed-forward 3DGS framework that incrementally maintains a persistent geometry-grounded scene state.<n>Experiments show that StreamSplat remains competitive with state-of-the-art feed-forward 3DGS methods under sparse causal inputs.<n>StreamSplat scales to long input streams with 256, 512, and 1024 views where fixed-view baselines run out of memory.
  arXiv  Detail & Related papers  (2026-08-03T03:48:20Z)
• Hidden Decoding at Scale: Latent Computation Scaling for Large Language Models [75.80275843320511]
  We study whether an existing backbone can keep improving by allocating more computation to each token while leaving the Transformer backbone fixed.<n>We propose Hidden Decoding, a sequence-length scaling method applied during continued pretraining.
  arXiv  Detail & Related papers  (2026-07-09T07:37:59Z)
• Anchor3R: Streaming 3D Reconstruction with Transient Anchors for Long-Horizon Visual Mapping [28.6268034980389]
  Long-horizon online visual mapping is a core capability for robot perception.<n>Recent feed-forward 3D reconstruction models predict poses in a fixed coordinate system tied to the first frame or a persistent scene memory.<n>We propose emphAnchor3R, a streaming 3D reconstruction framework that treats feed-forward reconstruction as current-centric local measurement prediction.
  arXiv  Detail & Related papers  (2026-06-03T16:00:13Z)
• HorizonStream: Long-Horizon Attention for Streaming 3D Reconstruction [18.749241400724493]
  Existing methods often suffer from drift, jitter, or collapse on long sequences.<n>We propose HorizonStream, a long-horizon Transformer that explicitly factorizes this kernel.<n>Experiments show that HorizonStream generalizes stably to sequences exceeding 10,000 frames with constant memory and linear time.
  arXiv  Detail & Related papers  (2026-05-22T17:50:48Z)
• DySink: Dynamic Frame Sinks for Autoregressive Long Video Generation [51.58559206569209]
  We propose DySink, a retrieval-based framework that maintains a compact memory bank and selects visually relevant historical frames as dynamic frame sinks.<n> Experiments on minute-long videos show that DySink consistently improves dynamic degree over strong baselines while also achieving higher temporal quality.
  arXiv  Detail & Related papers  (2026-05-20T11:01:01Z)
• Geometric Context Transformer for Streaming 3D Reconstruction [51.19524805829903]
  LingBot-Map is a feed-forward 3D foundation model for reconstructing scenes from streaming data.<n>A defining aspect of LingBot-Map lies in its carefully designed attention mechanism, which integrates an anchor context, a pose-reference window, and a trajectory memory.<n>This design keeps the streaming state compact while retaining rich geometric context, enabling stable efficient inference at around 20 FPS on 518 x 378 resolution inputs.
  arXiv  Detail & Related papers  (2026-04-15T17:58:13Z)
• Fast-SAM3D: 3Dfy Anything in Images but Faster [65.17322167628367]
  SAM3D enables scalable, open-world 3D reconstruction from complex scenes, yet its deployment is hindered by prohibitive inference latency.<n>We present textbfFast-SAM3D, a training-free framework that aligns computation with instantaneous generation complexity.
  arXiv  Detail & Related papers  (2026-02-05T04:27:59Z)
• LoL: Longer than Longer, Scaling Video Generation to Hour [50.945885467651216]
  This work achieves the first demonstration of real-time, streaming, and infinite-length video generation with little quality decay.<n>As an illustration, we generate continuous videos up to 12 hours in length, which, to our knowledge, is among the longest publicly demonstrated results in streaming video generation.
  arXiv  Detail & Related papers  (2026-01-23T17:21:35Z)
• InfiniteVGGT: Visual Geometry Grounded Transformer for Endless Streams [22.277387867568834]
  InfiniteVGGT is a causal visual geometry transformer that operationalizes the concept of a rolling memory through a bounded yet adaptive and perpetually expressive KV cache.<n>We introduce the Long3D benchmark, which, for the first time, enables a rigorous evaluation of continuous 3D geometry estimation on sequences about 10,000 frames.
  arXiv  Detail & Related papers  (2026-01-05T17:11:00Z)
• TTT3R: 3D Reconstruction as Test-Time Training [69.51086319339662]
  We revisit the 3D reconstruction foundation models from a Test-Time Training perspective.<n>We leverage the alignment confidence between the memory state and incoming observations to derive a closed-form learning rate.<n>This training-free intervention, termed TTT3R, substantially improves length generalization.
  arXiv  Detail & Related papers  (2025-09-30T17:59:51Z)
• Mixture of Contexts for Long Video Generation [72.96361488755986]
  We recast long-context video generation as an internal information retrieval task.<n>We propose a simple, learnable sparse attention routing module, Mixture of Contexts (MoC), as an effective long-term memory retrieval engine.<n>As we scale the data and gradually sparsify the routing, the model allocates compute to salient history, preserving identities, actions, and scenes over minutes of content.
  arXiv  Detail & Related papers  (2025-08-28T17:57:55Z)
• STream3R: Scalable Sequential 3D Reconstruction with Causal Transformer [72.88105562624838]
  We present STream3R, a novel approach to 3D reconstruction that reformulates pointmap prediction as a decoder-only Transformer problem.<n>By learning geometric priors from large-scale 3D datasets, STream3R generalizes well to diverse and challenging scenarios.<n>Our results underscore the potential of causal Transformer models for online 3D perception, paving the way for real-time 3D understanding in streaming environments.
  arXiv  Detail & Related papers  (2025-08-14T17:58:05Z)
• LONG3R: Long Sequence Streaming 3D Reconstruction [29.79885827038617]
  Long3R is a novel model designed for streaming multi-view 3D scene reconstruction over longer sequences.<n>Our model achieves real-time processing by operating recurrently, maintaining and updating memory with each new observation.<n>Experiments demonstrate that LONG3R outperforms state-of-the-art streaming methods, particularly for longer sequences.
  arXiv  Detail & Related papers  (2025-07-24T09:55:20Z)
• SplatVoxel: History-Aware Novel View Streaming without Temporal Training [29.759664150610362]
  We study the problem of novel view streaming from sparse-view videos.<n>Existing novel view synthesis methods struggle with temporal coherence and visual fidelity.<n>We propose a hybrid splat-voxel feed-forward scene reconstruction approach.
  arXiv  Detail & Related papers  (2025-03-18T20:00:47Z)
• Temporally Consistent Transformers for Video Generation [80.45230642225913]
  To generate accurate videos, algorithms have to understand the spatial and temporal dependencies in the world. No established benchmarks on complex data exist for rigorously evaluating video generation with long temporal dependencies. We introduce the Temporally Consistent Transformer (TECO), a generative model that substantially improves long-term consistency while also reducing sampling time.
  arXiv  Detail & Related papers  (2022-10-05T17:15:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.