mlsanityData to 7 Oct 2026

Alligat0R: Pre-Training Through Co-Visibility Segmentation for Relative Camera Pose Regression

arXiv 2503.07561 ↗First posted 10 Mar 2025cs.CVcited by 10

Not on the map. Why?

Abstract (arXiv)

Pre-training techniques have greatly advanced computer vision, with CroCo's cross-view completion approach yielding impressive results in tasks like 3D reconstruction and pose regression. However, cross-view completion is ill-posed in non-covisible regions, limiting its effectiveness. We introduce Alligat0R, a novel pre-training approach that replaces cross-view learning with a covisibility segmentation task. Our method predicts whether each pixel in one image is covisible in the second image, occluded, or outside the field of view, making the pre-training effective in both covisible and non-covisible regions, and provides interpretable predictions. To support this, we present Cub3, a large-scale dataset with 5M image pairs and dense covisibility annotations derived from the nuScenes and ScanNet datasets. Cub3 includes diverse scenarios with varying degrees of overlap. The experiments show that our novel pre-training method Alligat0R significantly outperforms CroCo in relative pose regression. Code is available at https://github.com/thibautloiseau/alligat0r.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 52 of its 63 references that are in the map. First posted on arXiv 10 Mar 2025.

Builds on

52 of 63 references.

References · 52

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  3. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  4. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  5. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  6. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  7. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  8. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  9. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  10. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,360
  11. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,526
  12. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
38 more
  1. 2024Depth Anything V2cited by 1,385
  2. 2017Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semanticscited by 1,345
  3. 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,313
  4. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  5. 2024Grounding Image Matching in 3D with MASt3Rcited by 960
  6. 2023LightGlue: Local Feature Matching at Light Speedcited by 583
  7. 2025Continuous 3D Perception Model with Persistent Statecited by 527
  8. 2024MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motioncited by 431
  9. 2024UniDepth: Universal Monocular Metric Depth Estimationcited by 375
  10. 2025Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Passcited by 356
  11. 2020DISK: Learning local features with policy gradientcited by 310
  12. 2023RoMa: Robust Dense Feature Matchingcited by 296
  13. 20243D Reconstruction with Spatial Memorycited by 269
  14. 2024No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Imagescited by 241
  15. 2021COTR: Correspondence Transformer for Matching Across Imagescited by 182
  16. 2022DKM: Dense Kernelized Feature Matching for Geometry Estimationcited by 174
  17. 2024Splatt3R: Zero-shot Gaussian Splatting from Uncalibrated Image Pairscited by 171
  18. 2022ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformercited by 170
  19. 2022CroCo v2: Improved Cross-view Completion Pre-training for Stereo Matching and Optical Flowcited by 165
  20. 2022CroCo: Self-Supervised Pre-training for 3D Vision Tasks by Cross-View Completioncited by 158
  21. 2022QuadTree Attention for Vision Transformerscited by 129
  22. 2022MatchFormer: Interleaving Attention in Transformers for Feature Matchingcited by 126
  23. 2023ALIKED: A Lighter Keypoint and Descriptor Extraction Network via Deformable Transformationcited by 120
  24. 2024Efficient LoFTR: Semi-Dense Local Feature Matching with Sparse-Like Speedcited by 119
  25. 2024XFeat: Accelerated Features for Lightweight Image Matchingcited by 113
  26. 2024Stereo4D: Learning How Things Move in 3D from Internet Stereo Videoscited by 89
  27. 2024Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localizationcited by 81
  28. 2021PDC-Net+: Enhanced Probabilistic Dense Correspondence Networkcited by 80
  29. 2023SiLK -- Simple Learned Keypointscited by 54
  30. 2023Compact Transformer Tracker with Correlative Masked Modelingcited by 50
  31. 2025Light3R-SfM: Towards Feed-forward Structure-from-Motioncited by 47
  32. 2024DeDoDe v2: Analyzing and Improving the DeDoDe Keypoint Detectorcited by 38
  33. 2024Cross-View Completion Models are Zero-shot Correspondence Estimatorscited by 30
  34. 2023PMatch: Paired Masked Image Modeling for Dense Geometric Matchingcited by 29
  35. 2024Masked Image Modeling: A Surveycited by 28
  36. 2022ECO-TR: Efficient Correspondences Finding Via Coarse-to-Fine Refinementcited by 18
  37. 2023End-to-End (Instance)-Image Goal Navigation through Correspondence as an Emergent Phenomenoncited by 16
  38. 2024Learning to Make Keypoints Sub-Pixel Accuratecited by 13

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 7; 2026: 3.

10 papers.

  1. 2025TTT3R: 3D Reconstruction as Test-Time Trainingcited by 122
  2. 2025AMB3R: Accurate Feed-forward Metric-scale 3D Reconstruction with Backendcited by 16
  3. 2025Deep Learning Reforms Image Matching: A Survey and Outlookcited by 7
  4. 2025MuM: Multi-View Masked Image Modeling for 3D Visioncited by 4
  5. 2025Co-VisiON: Co-Visibility ReasONing on Sparse Image Sets of Indoor Scenescited by 2
  6. 2025ShelfAware: Real-Time Semantic Localization in Quasi-Static Environments with Low-Cost Sensorscited by 1
  7. 2026Revisiting Cross-View Completion: Self-Supervised Pre-Training via Reconstruction Error Comparisoncited by 1
  8. 2025Audio-Visual Camera Pose Estimation with Passive Scene Sounds and In-the-Wild Videocited by 0
  9. 2026GIST: Multimodal Knowledge Extraction and Spatial Grounding via Intelligent Semantic Topologycited by 0
  10. 2026Emergent Multi-View Geometry Through Self-Distillationcited by 0