Jamie ShottonBen Glocker,2025-02-112013SHOTTON, Jamie, et al. Scene coordinate regression forests for camera relocalization in RGB-D images. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2013. p. 2930-2937.‏https://openaccess.thecvf.com/content_cvpr_2013/html/Shotton_Scene_Coordinate_Regression_2013_CVPR_paper.htmlhttps://ds.uofallujah.edu.iq/handle/123456789/385This paper presents a method for estimating the pose of an RGB-D camera relative to a known 3D scene using a single image. The approach employs a regression forest to predict 3D correspondences for each pixel without requiring feature descriptors or interest point detectors. A robust optimization process, including preemptive RANSAC, refines the camera pose based on inliers. The method is evaluated on various scenes and outperforms state-of-the-art baselines in accuracy and relocalization performance.We address the problem of inferring the pose of an RGB-D camera relative to a known 3D scene, given only a single acquired image. Our approach employs a regres sion forest that is capable of inferring an estimate of each pixel’s correspondence to 3D points in the scene’s world coordinate frame. The forest uses only simple depth and RGB pixel comparison features, and does not require the computation of feature descriptors. The forest is trained to be capable of predicting correspondences at any pixel, so no interest point detectors are required. The camera pose is inferred using a robust optimization scheme. This starts with an initial set of hypothesized camera poses, con structed by applying the forest at a small fraction of image pixels. Preemptive RANSAC then iterates sampling more pixels at which to evaluate the forest, counting inliers, and refining the hypothesized poses. We evaluate on several var ied scenes captured with an RGB-D camera and observe that the proposed technique achieves highly accurate relo calization and substantially out-performs two state of the art baselines.enArticle