Learning of low-level feature keypoints for accurate and robust detection

Suwanwimolkul, Suwichaya; Komorita, Satoshi; Tasaka, Kazuyuki

doi:10.1109/wacv48630.2021.00231

Cited by 17 publications

(3 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Suwanwimolkul et al . [8] noted that since the keypoint selection is more handcrafted in approaches like D2-Net [4] and ASLFeat [5], there are no guarantees that the chosen keypoints can match the learned descriptors. Consequently, the accuracy of the matched keypoints is not usually very high.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Improving the matching of deformable objects by learning to detect keypoints

Cadar,

Melo,

Kanagasabapathi

et al. 2023

Pattern Recognition Letters

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

“…Recently, learning-based systems [1,2,3,4,5,6,7,8,9] have become more prevalent in feature detection techniques, producing outcomes that greatly exceed handcrafted keypoint detectors [10,11]. Despite the advances in learn-ing feature representations, end-to-end learning of keypoint detection is still a challenging problem.…”

Section: Introductionmentioning

confidence: 99%

Improving the matching of deformable objects by learning to detect keypoints

Cadar,

Melo,

Kanagasabapathi

et al. 2023

Pattern Recognition Letters

View full text Add to dashboard Cite

“…R2D2 [28] computes the repeatability and reliability maps for keypoint detection, and it trains the descriptors with AP loss. Suwichaya recently added a low-level feature LLF detector to the R2D2 to improve keypoint accuracy [29]. DISK [22] uses reinforcement learning to train the score map and descriptor map.…”

Section: B Joint Keypoint and Descriptor Learningmentioning

confidence: 99%

ALIKED: A Lighter Keypoint and Descriptor Extraction Network via Deformable Transformation

Zhao

Chen

et al. 2023

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

Image keypoints and descriptors play a crucial role in many visual measurement tasks. In recent years, deep neural networks have been widely used to improve the performance of keypoint and descriptor extraction. However, the conventional convolution operations do not provide the geometric invariance required for the descriptor. To address this issue, we propose the Sparse Deformable Descriptor Head (SDDH), which learns the deformable positions of supporting features for each keypoint and constructs deformable descriptors. Furthermore, SDDH extracts descriptors at sparse keypoints instead of a dense descriptor map, which enables efficient extraction of descriptors with strong expressiveness. In addition, we relax the neural reprojection error (NRE) loss from dense to sparse to train the extracted sparse descriptors. Experimental results show that the proposed network is both efficient and powerful in various visual measurement tasks, including image matching, 3D reconstruction, and visual relocalization.

show abstract