Slot3R: Set-Associative Spatial Memory for Streaming 3D Reconstruction

arXiv:2610.12282 · cs.CV · Submitted 2026-10-08 · Read on arXiv

cs.CV

Submitted: 2026-10-08

Updated: 2026-10-08

Project page: https://ashleyxyz.github.io/Slot-3R/Abstract

Terminology

• Core — 34× in this paper · explained in 2 episodes
• Ablation — 10× in this paper · explained in 1 episode
• Transformer — 9× in this paper · explained in 2 episodes
• Alignment — 8× in this paper · explained in 1 episode
• Benchmark — 7× in this paper · explained in 2 episodes
• Accuracy — 4× in this paper · explained in 1 episode
• Cosine Similarity — 4× in this paper · explained in 3 episodes
• Consistency — 3× in this paper · explained in 3 episodes
• Associative Memory — 2× in this paper · explained in 2 episodes
• Fine-Tuning — 2× in this paper · explained in 5 episodes
• Pruning — 2× in this paper · explained in 4 episodes
• Ablation study — 1× in this paper · explained in 1 episode
• Calibration — 1× in this paper · explained in 7 episodes
• Optical Flow — 1× in this paper · explained in 2 episodes
• Quantization — 1× in this paper · explained in 12 episodes

Sources

• Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass
• $\pi^3$: Permutation-Equivariant Visual Geometry Learning
• 3D Reconstruction with Spatial Memory
• Point3R: Streaming 3D Reconstruction with Explicit Spatial Pointer Memory
• Continuous 3D Perception Model with Persistent State
• Rethinking the State Update Gate for Long-Sequence Recurrent 3D Reconstruction
• InfiniteVGGT: Visual Geometry Grounded Transformer for Endless Streams
• Attention Itself Could Retrieve.RetrieveVGGT: Training-Free Long Context Streaming 3D Reconstruction via Query-Key Similarity Retrieval
• HD-VGGT: High-Resolution Visual Geometry Transformer
• VGGT-Long: Chunk it, Loop it, Align it -- Pushing VGGT's Limits on Kilometer-scale Long RGB Sequences
• Diversity-aware View Partitioning for Scalable VGGT
• FastVGGT: Training-Free Acceleration of Visual Geometry Transformer
• Scal3R: Scalable Test-Time Training for Large-Scale 3D Reconstruction
• ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Training
• Efficiently Reconstructing Dynamic Scenes One D4RT at a Time
• SM4RT: Learning Structured Motion Geometry for 4D Reconstruction
• TTT3R: 3D Reconstruction as Test-Time Training
• MeMix: Writing Less, Remembering More for Streaming 3D Reconstruction
• FILT3R: Latent State Adaptive Kalman Filter for Streaming 3D Reconstruction
• Mamba-VGGT: Persistent Long-Sequence Video Geometry Grounded Transformer via External Sliding Window Mamba Memory

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models