Sökning: WFRF:(Edstedt Johan) >
DeDoDe: Detect, Don...
DeDoDe: Detect, Don't Describe - Describe, Don't Detect for Local Feature Matching
-
Edstedt, Johan (författare)
-
- Bökman, Georg, 1994 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
Wadenbäck, Mårten (författare)
-
visa fler...
-
Felsberg, Michael (författare)
-
visa färre...
-
(creator_code:org_t)
- 2024
- 2024
- Engelska.
-
Ingår i: Proceedings - 2024 International Conference on 3D Vision, 3DV 2024. ; , s. 148-157
- Relaterad länk:
-
https://doi.org/10.1...
-
visa fler...
-
https://research.cha...
-
visa färre...
Abstract
Ämnesord
Stäng
- Keypoint detection is a pivotal step in 3D reconstruction, whereby sets of (up to) K points are detected in each view of a scene. Crucially, the detected points need to be consistent between views, i.e., correspond to the same 3D point in the scene. One of the main challenges with keypoint detection is the formulation of the learning objective. Previous learning-based methods typically jointly learn descriptors with keypoints, and treat the keypoint detection as a binary classification task on mutual nearest neighbours. However, basing keypoint detection on descriptor nearest neighbours is a proxy task, which is not guaranteed to produce 3D-consistent keypoints. Furthermore, this ties the keypoints to a specific descriptor, complicating downstream usage. In this work, we instead learn keypoints directly from 3D consistency. To this end, we train the detector to detect tracks from large-scale SfM. As these points are often overly sparse, we derive a semi-supervised two-view detection objective to expand this set to a desired number of detections. To train a descriptor, we maximize the mutual nearest neighbour objective over the keypoints with a separate network. Results show that our approach, DeDoDe, achieves significant gains on multiple geometry benchmarks. Code is provided at http://github.com/Parskatt/DeDoDegithub.com/Parskatt/DeDoDe.
Ämnesord
- TEKNIK OCH TEKNOLOGIER -- Elektroteknik och elektronik -- Robotteknik och automation (hsv//swe)
- ENGINEERING AND TECHNOLOGY -- Electrical Engineering, Electronic Engineering, Information Engineering -- Robotics (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
Nyckelord
- feature matching
- structure-from-motion
- keypoint detection
- 3D reconstruction
- local feature matching
- image matching
Publikations- och innehållstyp
- kon (ämneskategori)
- ref (ämneskategori)