cs.CVNov 3, 2026

Mamba3R: Fixed-Capacity Selective State Memory for Long-Sequence 3D Reconstruction

Authors: Zhenxiang Xiong, Tianchen Deng, Nailin Wang, Lei Zhao, Guole Shen, Weidong Chen

Organizations: The authors are with the Institute of Medical Robotics and the School of Automation and Intelligent Sensing, Shanghai Jiao Tong University, Shanghai 200240, China, and also with the Key Laboratory of System Control and Information Processing, Ministry of Education, Shanghai 200240, China

Abstract

Streaming 3D reconstruction must maintain geometric quality over long inputs within a fixed GPU budget. A pretrained recurrent model provides bounded state, but freezing its backbone also fixes its learned state-observation interactions. We introduce Mamba3R, a jointly designed state-memory adapter that provides a trainable adaptation path without updating the backbone or task heads. A selective state-space mixer jointly transforms native state features and a fixed-capacity auxiliary memory, explicit read and write paths carry information across frames, and a decomposed feedback path regulates what the added memory feeds back. Only 1.28 million adapter parameters are trained. Under a shared periodic-reset protocol, it lowers trajectory error and improves accumulated surface and depth accuracy over recurrent baselines on public indoor and outdoor datasets and robot-recorded traversals, gaining up to 40 percent on one metric. Matched comparisons support both the choice of the mixing operator and the utility of carrying the auxiliary memory across frames. Allocated memory remains unchanged and throughput moves by at most 1.4 percent from one to two thousand frames. Replaying its recorded traversals offline, the platform's RTX 4060 averages 11.7 frames per second within 5.1 GiB with the same checkpoint, a mean throughput covering the 10 Hz sensor rate.

Figures & tables

Explore similar work

CardsList
  1. Mem3R: Streaming 3D Reconstruction with Hybrid Memory

    Apr 8, 2026Changkun Liu, Jiezhi Yang, Zeman Li +4Streaming 3D ReconstructionMemory-Augmented Neural Networks

  2. TTSA3R: Training-Free Temporal-Spatial Adaptive Persistent State for Streaming 3D Reconstruction

    Jan 30, 2026Zhijie Zheng, Xinhao Xiang, Jiawei ZhangStreaming 3D Reconstruction3D Reconstruction

  3. BAT3R: Robust Online 3D Reconstruction with Bayesian Adaptive State Updates

    Oct 12, 2026Lijian Li, Yuanpeng He, Yudian Zheng +6Streaming 3D ReconstructionBayesian Filtering