Abstract
This paper proposes a keypoint regressor (KeyReg), which consists of multi-layer random forest (MRF) regressor and single random forest (SRF) classifier modules. To increase the keypoints’ repeatability, the MRF regressor is applied to multi-scale images in a shared rules manner, and keypoints predicted at each scale are given a confidence score through the SRF for reliability measurement. Each candidate point is detected as the final keypoint through a non-maxima suppression process based on a confidence score. The MRF structure of KeyReg is designed to maintain a coarse-to-fine structure by varying the number of nodes per layer. In addition, the accuracy of the matching can be improved by removing less confidential keypoints through the continuous SRF classifier. KeyReg is the first approach to apply an MRF to the keypoint regression and is designed to run on a CPU rather than a GPU compared to DNN-based approaches. KeyReg training was conducted using COCO, and positive and negative examples were automatically obtained under a self-supervised learning method between the original image and a warped image. The proposed KeyReg showed superior performance in terms of repeatability, the accuracy of the homography, mean matching accuracy (MMA), and localization errors on HPatches dataset compared to state-of-the-art methods.
Highlights
Image registration is a process of transforming different images of a scene into the same coordinate system
Because using only one global regressor does not cope with an overfitting or unseen homography transformations during the training process [30], in this study, we propose an multi-layer random forest (MRF) regressor that can apply a sequential regression with a coarse-to-fine pipeline structure inspired by [31], [32] and end-to-end learning to reduce overfitting and design a model that is robust to image warping
HPATCHES HOMOGRAPHY ESTIMATION To evaluate the performance of keypoint regressor (KeyReg) on the matching ability on the HPatches dataset, we evaluated KeyReg against handcrafted approaches as well as state-of-the-art convolutional neural network (CNN)-based detectors in terms of the homography accuracy and mean matching accuracy (MMA)
Summary
Image registration is a process of transforming different images of a scene into the same coordinate system. (c) A feature descriptor is extracted from the final predicted keypoint and entered into the SRF module to calculate the confidence score of the keypoint. KeyReg is similar to [23] in that it uses a handcrafted feature in combination with a classifier, but KeyReg improves the performance of keypoint matching by combining MRF and SRF that enable self-supervised learning instead of CNN.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
