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

arXiv:2608.00925v1 Announce Type: new Abstract: 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 extbf{100%} sequence success (extbf{125/125}) under the stated criterion and the lowest ATE among the evaluated methods on all five benchmarks, reducing ATE by extbf{30--88%} relative to the best ERP-native baseline on each benchmark.

Source: arXiv cs.CV | 2026-08-04