# Long-Term Multi-Session 3D Reconstruction Under Substantial Appearance Change

arXiv: https://arxiv.org/abs/2602.20584 · HTML: https://mlsanity.com/paper/2602.20584/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 66 references that are in the map.
- First posted on arXiv 24 Feb 2026.

## Builds on (most cited first, 25 of 25)

- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [SuperPoint: Self-Supervised Interest Point Detection and Description](https://mlsanity.com/paper/1712.07629/) (2017)
- [LightGlue: Local Feature Matching at Light Speed](https://mlsanity.com/paper/2306.13643/) (2023)
- [MapAnything: Universal Feed-Forward Metric 3D Reconstruction](https://mlsanity.com/paper/2509.13414/) (2025)
- [KISS-ICP: In Defense of Point-to-Point ICP -- Simple, Accurate, and Robust Registration If Done the Right Way](https://mlsanity.com/paper/2209.15397/) (2022)
- [A comprehensive survey on point cloud registration](https://mlsanity.com/paper/2103.02690/) (2021)
- [MegaLoc: One Retrieval to Place Them All](https://mlsanity.com/paper/2502.17237/) (2025)
- [Khronos: A Unified Approach for Spatio-Temporal Metric-Semantic SLAM in Dynamic Environments](https://mlsanity.com/paper/2402.13817/) (2024)
- [Real-Time Dense 3D Mapping of Underwater Environments](https://mlsanity.com/paper/2304.02704/) (2023)
- [Point Label Aware Superpixels for Multi-species Segmentation of Underwater Imagery](https://mlsanity.com/paper/2202.13487/) (2022)
- [High Definition, Inexpensive, Underwater Mapping](https://mlsanity.com/paper/2203.05640/) (2022)
- [Scalable Semantic 3D Mapping of Coral Reefs with Deep Learning](https://mlsanity.com/paper/2309.12804/) (2023)
- [ObVi-SLAM: Long-Term Object-Visual SLAM](https://mlsanity.com/paper/2309.15268/) (2023)
- [BUFFER-X: Towards Zero-Shot Point Cloud Registration in Diverse Scenes](https://mlsanity.com/paper/2503.07940/) (2025)
- [Image Labels Are All You Need for Coarse Seagrass Segmentation](https://mlsanity.com/paper/2303.00973/) (2023)
- [Eiffel Tower: A Deep-Sea Underwater Dataset for Long-Term Visual Localization](https://mlsanity.com/paper/2305.05301/) (2023)
- [TURTLMap: Real-time Localization and Dense Mapping of Low-texture Underwater Environments with a Low-cost Unmanned Underwater Vehicle](https://mlsanity.com/paper/2408.01569/) (2024)
- [Refractive COLMAP: Refractive Structure-from-Motion Revisited](https://mlsanity.com/paper/2403.08640/) (2024)
- [ICP Algorithm: Theory, Practice And Its SLAM-oriented Taxonomy](https://mlsanity.com/paper/2206.06435/) (2022)
- [Semihierarchical Reconstruction and Weak-area Revisiting for Robotic Visual Seafloor Mapping](https://mlsanity.com/paper/2308.06147/) (2023)
- [Reducing Label Dependency for Underwater Scene Understanding: A Survey of Datasets, Techniques and Applications](https://mlsanity.com/paper/2411.11287/) (2024)
- [Image-Based Relocalization and Alignment for Long-Term Monitoring of Dynamic Underwater Environments](https://mlsanity.com/paper/2503.04096/) (2025)
- [Going Places: Place Recognition in Artificial and Natural Systems](https://mlsanity.com/paper/2511.14341/) (2025)
- [Long-Term Online Multi-Session Graph-Based SPLAM with Memory Management](https://mlsanity.com/paper/2301.00050/) (2022)
- [Online Refractive Camera Model Calibration in Visual Inertial Odometry](https://mlsanity.com/paper/2409.12074/) (2024)

## Abstract (arXiv)

Long-term environmental monitoring requires the ability to reconstruct and align 3D models across repeated site visits separated by months or years. However, existing Structure-from-Motion (SfM) pipelines implicitly assume near-simultaneous image capture and limited appearance change, and therefore fail when applied to long-term monitoring scenarios such as coral reef surveys, where substantial visual and structural change is common. In this paper, we show that the primary limitation of current approaches lies in their reliance on post-hoc alignment of independently reconstructed sessions, which is insufficient under large temporal appearance change. We address this limitation by enforcing cross-session correspondences directly within a joint SfM reconstruction. Our approach combines complementary handcrafted and learned visual features to robustly establish correspondences across large temporal gaps, enabling the reconstruction of a single coherent 3D model from imagery captured years apart, where standard independent and joint SfM pipelines break down. We evaluate our method on long-term coral reef datasets exhibiting significant real-world change, and demonstrate consistent joint reconstruction across sessions in cases where existing methods fail to produce coherent reconstructions. To ensure scalability to large datasets, we further restrict expensive learned feature matching to a small set of likely cross-session image pairs identified via visual place recognition, which reduces computational cost and improves alignment robustness.