Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2601.12423 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911383867621376 |
|---|---|
| author | Clerc, Antonin Quellmalz, Michael Piening, Moritz Flotho, Philipp Kornhardt, Gregor Steidl, Gabriele |
| author_facet | Clerc, Antonin Quellmalz, Michael Piening, Moritz Flotho, Philipp Kornhardt, Gregor Steidl, Gabriele |
| contents | Stereo vision between images faces a range of challenges, including occlusions, motion, and camera distortions, across applications in autonomous driving, robotics, and face analysis. Due to parameter sensitivity, further complications arise for stereo matching with sparse features, such as facial landmarks. To overcome this ill-posedness and enable unsupervised sparse matching, we consider line constraints of the camera geometry from an optimal transport (OT) viewpoint. Formulating camera-projected points as (half)lines, we propose the use of the classical epipolar distance as well as a 3D ray distance to quantify matching quality. Employing these distances as a cost function of a (partial) OT problem, we arrive at efficiently solvable assignment problems. Moreover, we extend our approach to unsupervised object matching by formulating it as a hierarchical OT problem. The resulting algorithms allow for efficient feature and object matching, as demonstrated in our numerical experiments. Here, we focus on applications in facial analysis, where we aim to match distinct landmarking conventions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_12423 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HOT-POT: Optimal Transport for Sparse Stereo Matching Clerc, Antonin Quellmalz, Michael Piening, Moritz Flotho, Philipp Kornhardt, Gregor Steidl, Gabriele Computer Vision and Pattern Recognition Optimization and Control 68U10, 49Q22, 90C46 Stereo vision between images faces a range of challenges, including occlusions, motion, and camera distortions, across applications in autonomous driving, robotics, and face analysis. Due to parameter sensitivity, further complications arise for stereo matching with sparse features, such as facial landmarks. To overcome this ill-posedness and enable unsupervised sparse matching, we consider line constraints of the camera geometry from an optimal transport (OT) viewpoint. Formulating camera-projected points as (half)lines, we propose the use of the classical epipolar distance as well as a 3D ray distance to quantify matching quality. Employing these distances as a cost function of a (partial) OT problem, we arrive at efficiently solvable assignment problems. Moreover, we extend our approach to unsupervised object matching by formulating it as a hierarchical OT problem. The resulting algorithms allow for efficient feature and object matching, as demonstrated in our numerical experiments. Here, we focus on applications in facial analysis, where we aim to match distinct landmarking conventions. |
| title | HOT-POT: Optimal Transport for Sparse Stereo Matching |
| topic | Computer Vision and Pattern Recognition Optimization and Control 68U10, 49Q22, 90C46 |
| url | https://arxiv.org/abs/2601.12423 |