mlsanityData to 7 Oct 2026

Mismatched: Evaluating the Limits of Image Matching Approaches and Benchmarks

arXiv 2408.16445 ↗First posted 29 Aug 2024cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

Three-dimensional (3D) reconstruction from two-dimensional images is an active research field in computer vision, with applications ranging from navigation and object tracking to segmentation and three-dimensional modeling. Traditionally, parametric techniques have been employed for this task. However, recent advancements have seen a shift towards learning-based methods. Given the rapid pace of research and the frequent introduction of new image matching methods, it is essential to evaluate them. In this paper, we present a comprehensive evaluation of various image matching methods using a structure-from-motion pipeline. We assess the performance of these methods on both in-domain and out-of-domain datasets, identifying key limitations in both the methods and benchmarks. We also investigate the impact of edge detection as a pre-processing step. Our analysis reveals that image matching for 3D reconstruction remains an open challenge, necessitating careful selection and tuning of models for specific scenarios, while also highlighting mismatches in how metrics currently represent method performance.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 31 of its 60 references that are in the map. First posted on arXiv 29 Aug 2024.

Builds on

31 of 60 references.

References · 31

  1. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  2. 2021Vision Transformers for Dense Predictioncited by 1,789
  3. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,526
  4. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  5. 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,313
  6. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  7. 2024Grounding Image Matching in 3D with MASt3Rcited by 960
  8. 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 768
  9. 2023LightGlue: Local Feature Matching at Light Speedcited by 583
  10. 2020DISK: Learning local features with policy gradientcited by 310
  11. 2016LIFT: Learned Invariant Feature Transformcited by 305
  12. 2023RoMa: Robust Dense Feature Matchingcited by 296
19 more
  1. 2020Image Matching across Wide Baselines: From Paper to Practicecited by 224
  2. 2019MOTS: Multi-Object Tracking and Segmentationcited by 222
  3. 2022CroCo: Self-Supervised Pre-training for 3D Vision Tasks by Cross-View Completioncited by 158
  4. 2023Optimal Transport Aggregation for Visual Place Recognitioncited by 151
  5. 2023ALIKED: A Lighter Keypoint and Descriptor Extraction Network via Deformable Transformationcited by 120
  6. 2024Efficient LoFTR: Semi-Dense Local Feature Matching with Sparse-Like Speedcited by 119
  7. 2024XFeat: Accelerated Features for Lightweight Image Matchingcited by 113
  8. 2022Map-free Visual Relocalization: Metric Pose Relative to a Single Imagecited by 112
  9. 2019Dense Extreme Inception Network: Towards a Robust CNN Model for Edge Detectioncited by 104
  10. 2021ALIKE: Accurate and Lightweight Keypoint Detection and Descriptor Extractioncited by 77
  11. 2024GIM: Learning Generalizable Image Matcher From Internet Videoscited by 72
  12. 2023DeDoDe: Detect, Don't Describe -- Describe, Don't Detect for Local Feature Matchingcited by 62
  13. 2024OmniGlue: Generalizable Feature Matching with Foundation Model Guidancecited by 55
  14. 2021Dense Extreme Inception Network for Edge Detectioncited by 37
  15. 2024Local Feature Matching Using Deep Learning: A Surveycited by 26
  16. 2024Matching 2D Images in 3D: Metric Relative Pose from Metric Correspondencescited by 25
  17. 2023Steerers: A framework for rotation equivariant keypoint descriptorscited by 21
  18. 2024EarthMatch: Iterative Coregistration for Fine-grained Localization of Astronaut Photographycited by 13
  19. 2024Learning to Make Keypoints Sub-Pixel Accuratecited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 0; 2026: 3.

4 papers.

  1. 2024Robotic Arm Platform for Multi-View Image Acquisition and 3D Reconstruction in Minimally Invasive Surgerycited by 1
  2. 2026CM-Bench: A Comprehensive Cross-Modal Feature Matching Benchmark Bridging Visible and Infrared Imagescited by 1
  3. 2026Understanding and Optimizing Attention-Based Sparse Matching for Diverse Local Featurescited by 0
  4. 2026Accuracy potential of visual localization exploiting high-end street-level imagerycited by 0