# Look Up and Look Back: Hidden Attention and Latent Orientation in a Frozen Foundation Model for Panoramic SLAM

arXiv: https://arxiv.org/abs/2608.00925 · HTML: https://mlsanity.com/paper/2608.00925/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 42 references that are in the map.
- First posted on arXiv 2 Aug 2026.

## Builds on (most cited first, 19 of 19)

- [DUSt3R: Geometric 3D Vision Made Easy](https://mlsanity.com/paper/2312.14132/) (2023)
- [ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM](https://mlsanity.com/paper/2007.11898/) (2020)
- [DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras](https://mlsanity.com/paper/2108.10869/) (2021)
- [TartanAir: A Dataset to Push the Limits of Visual SLAM](https://mlsanity.com/paper/2003.14338/) (2020)
- [Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass](https://mlsanity.com/paper/2501.13928/) (2025)
- [MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors](https://mlsanity.com/paper/2412.12392/) (2024)
- [VGGT-SLAM: Dense RGB SLAM Optimized on the SL(4) Manifold](https://mlsanity.com/paper/2505.12549/) (2025)
- [VGGT-Long: Chunk it, Loop it, Align it -- Pushing VGGT's Limits on Kilometer-scale Long RGB Sequences](https://mlsanity.com/paper/2507.16443/) (2025)
- [SLAM3R: Real-Time Dense Scene Reconstruction from Monocular RGB Videos](https://mlsanity.com/paper/2412.09401/) (2024)
- [OpenVSLAM: A Versatile Visual SLAM Framework](https://mlsanity.com/paper/1910.01122/) (2019)
- [360Loc: A Dataset and Benchmark for Omnidirectional Visual Localization with Cross-device Queries](https://mlsanity.com/paper/2311.17389/) (2023)
- [VGGT-SLAM 2.0: Real-time Dense Feed-forward Scene Reconstruction](https://mlsanity.com/paper/2601.19887/) (2026)
- [CubemapSLAM: A Piecewise-Pinhole Monocular Fisheye SLAM System](https://mlsanity.com/paper/1811.12633/) (2018)
- [PanoVGGT: Feed-Forward 3D Reconstruction from Panoramic Imagery](https://mlsanity.com/paper/2603.17571/) (2026)
- [AirSim360: A Panoramic Simulation Platform within Drone View](https://mlsanity.com/paper/2512.02009/) (2025)
- [LF-VISLAM: A SLAM Framework for Large Field-of-View Cameras with Negative Imaging Plane on Mobile Agents](https://mlsanity.com/paper/2209.05167/) (2022)
- [Holo360D: A Large-Scale Real-World Dataset with Continuous Trajectories for Advancing Panoramic 3D Reconstruction and Beyond](https://mlsanity.com/paper/2604.22482/) (2026)
- [360DVO: Deep Visual Odometry for Monocular 360-Degree Camera](https://mlsanity.com/paper/2601.02309/) (2026)
- [Learning Projection-Aware 360-Degree Image Rectification via Dual-Projection Fusion](https://mlsanity.com/paper/2512.00911/) (2025)

## Built on by (most cited first, 1 of 1)

- [Many Eyes, One World: Feed-Forward 3D Reconstruction from Mixed Cameras](https://mlsanity.com/paper/2609.35658/) (2026)

## Abstract (arXiv)

Monocular panoramic SLAM benefits from substantial visual overlap under large camera rotations, yet remains prone to errors caused by camera tilt, scale drift, and false loop closures. We show that a frozen panoramic geometry foundation model provides useful internal cues beyond its explicit geometric outputs: intermediate tokens encode gravity in the camera frame, while cross-view attention provides a compatibility cue for potential revisits. Building on these cues, we present HALO-SLAM. A gravity readout enables IMU-free spherical upright canonicalization. For loop closure, we introduce a cost-aware three-stage cascade combining DBoW2 event-level retrieval, attention-based compatibility filtering, and dense geometric validation through symmetric submap augmentation. Accepted revisits yield pixel-aligned 3D--3D correspondences in both local gauges, from which robust Sim(3) constraints are estimated and jointly optimized with sequential constraints in a global pose graph. Across 125 sequences from five real-world panoramic benchmarks, our method achieves 100% sequence success (125/125) under the stated criterion and the lowest ATE among the evaluated methods on all five benchmarks, reducing ATE by 30--88% relative to the best ERP-native baseline on each benchmark.