Exploiting uncertainty in regression forests for accurate camera relocalization
The DOI matches its Crossref record; 23 of the 23 references it cites were found in a registry or an index.
Abstract
Recent advances in camera relocalization use predictions from a regression forest to guide the camera pose optimization procedure. In these methods, each tree associates one pixel with a point in the scene's 3D world coordinate frame. In previous work, these predictions were point estimates and the subsequent camera pose optimization implicitly assumed an isotropic distribution of these estimates. In this paper, we train a regression forest to predict mixtures of anisotropic 3D Gaussians and show how the predicted uncertainties can be taken into account for continuous pose optimization. Experiments show that our proposed method is able to relocalize up to 40% more frames than the state of the art.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Likewise it is possible to localize the 6DoF of a camera with regression forest [247]”
From Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016) · cites this paper as [247]
Citations
Cites
Mean shift: a robust approach toward feature space analysis
Comaniciu, D., & Meer, P. (2002) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.1000236 · Cited by 11,305 · cited 1×
“Similar to [19], we run mean shift mode detection [3] to find the modes of the empirical distribution in each leaf.”
MethodKinectFusion: Real-time dense surface mapping and tracking
Newcombe, R. A., Fitzgibbon, A., Izadi, S., Hilliges, O., Molyneaux, D., Kim, D., Davison, A. J., Kohi, P., Shotton, J., & Hodges, S. (2011) · 2011 10th IEEE International Symposium on Mixed and Augmented Reality · doi:10.1109/ismar.2011.6092378 · Cited by 4,008 · cited 1×
“Simultaneous Localization and Mapping (SLAM) systems such as [14] [15] [17] [7] and commercial systems such as HoloLens or Project Tango demonstrate that visual SLAM is a maturing technology.”
BackgroundSelf-citationLSD-SLAM: Large-scale direct monocular SLAM
Engel, J., Schöps, T., & Cremers, D. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10605-2_54 · Cited by 3,851 · cited 1×
“Simultaneous Localization and Mapping (SLAM) systems such as [14] [15] [17] [7] and commercial systems such as HoloLens or Project Tango demonstrate that visual SLAM is a maturing technology.”
BackgroundA solution for the best rotation to relate two sets of vectors
Kabsch, W. (1976) · Acta Crystallographica Section A · doi:10.1107/s0567739476001873 · Cited by 3,187 · cited 1×
“When per-pixel depth information is available, one can estimate the transformation parameters from the current view to the model with three local positions (current camera space) and their corresponding positions in the scene (world coordinates) using the Kabsch algorithm [11].”
MethodDTAM: Dense tracking and mapping in real-time
Newcombe, R. A., Lovegrove, S. J., & Davison, A. J. (2011) · 2011 International Conference on Computer Vision · doi:10.1109/iccv.2011.6126513 · Cited by 2,078 · cited 1×
“Simultaneous Localization and Mapping (SLAM) systems such as [14] [15] [17] [7] and commercial systems such as HoloLens or Project Tango demonstrate that visual SLAM is a maturing technology.”
BackgroundReal-time human pose recognition in parts from single depth images
Shotton, J., Sharp, T., Kipman, A., Fitzgibbon, A., Finocchio, M., Blake, A., Cook, M., & Moore, R. (2013) · Communications of the ACM · doi:10.1145/2398356.2398381 · Cited by 2,044 · cited 1×
“Following [19], we use features based on pairwise pixel relationships [13, 20] as they have proven to be well-suited for the task of camera relocalization.”
MethodSelf-citationFAB-MAP: Probabilistic Localization and Mapping in the Space of Appearance
Cummins, M., & Newman, P. (2008) · The International Journal of Robotics Research · doi:10.1177/0278364908090961 · Cited by 1,472 · cited 1×
“A few successful approaches also exist that do not belong to either of those two classes [5, 16].”
BackgroundReal-time 3D reconstruction at scale using voxel hashing
Nießner, M., Zollhöfer, M., Izadi, S., & Stamminger, M. (2013) · ACM Transactions on Graphics · doi:10.1145/2508363.2508374 · Cited by 1,002 · cited 1×
“Simultaneous Localization and Mapping (SLAM) systems such as [14] [15] [17] [7] and commercial systems such as HoloLens or Project Tango demonstrate that visual SLAM is a maturing technology.”
BackgroundSelf-citationScene Coordinate Regression Forests for Camera Relocalization in RGB-D Images
Shotton, J., Glocker, B., Zach, C., Izadi, S., Criminisi, A., & Fitzgibbon, A. (2013) · 2013 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2013.377 · Cited by 930 · cited 3×
“Perhaps the simplest approach, used in [19], is to fit an isotropic Gaussian to the labels in S.”
BackgroundMethodInfluentialSelf-citationChum, O., Matas, J., & Kittler, J. (2003) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-45243-0_31 · Cited by 843 · cited 3×
“Having computed the set of inliers, [2] suggests performing a local optimization step to refine the hypotheses.”
MethodInfluential
Cited by
Items 101–182 of 182
Structure-guided camera localization for indoor environments
Li, Q., Cao, R., Liu, K., Li, Z., Zhu, J., Bao, Z., Xu, F., Li, Q., Huang, X., & Qiu, G. (2023) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2023.05.034 · Cited by 16
Camera Relocalization using Deep Point Cloud Generation and Hand-crafted Feature Refinement
Junyi, W., & Qi, Y. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561289 · Cited by 16
“[31] trained regression forest to predict multi-model distributions of scene coordinates and show how predicted”
BackgroundCamera Relocalization by Exploiting Multi-View Constraints for Scene Coordinates Regression
Cai, M., Zhan, H., Weerasekera, C. S., Li, K., & Reid, I. D. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00469 · Cited by 16
“[25] exploits the uncertainty in the estimate from the Random Forest to benefit the pose optimization.”
MethodSupercharging Floorplan Localization with Semantic Rays
Grader, Y., & Averbuch‐Elor, H. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02517 · Cited by 15
“Camera localization is a longstanding problem in computer vision, with significant applications in 3D reconstruction [19, 25–28], augmented reality [11, 16, 23, 29, 31], and navigation [20, 24, 30, 32, 36].”
Regression Forest Based RGB-D Visual Relocalization Using Coarse-to-Fine Strategy
Wang, J., Wang, P., Dai, D., Xu, M., & Chen, Z. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.3000429 · Cited by 15
“To deal with the visual ambiguity, Valentin et al. [6] replace the modes used in the leaves of the forests with mixtures of anisotropic 3D Gaussians and optimize camera pose considering the predicted uncertainty, which greatly improve the relocalization performance.”
BackgroundResultTransCNNLoc: End-to-end pixel-level learning for 2D-to-3D pose estimation in dynamic indoor scenes
Tang, S., Li, Y., Wan, J., Li, Y., Zhou, B., Guo, R., Wang, W., & Feng, Y. (2023) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2023.12.006 · Cited by 14
Multi-task learning and joint refinement between camera localization and object detection
Wang, J., & Qi, Y. (2024) · Computational Visual Media · doi:10.1007/s41095-022-0319-z · Cited by 14
State Estimation for Hybrid Wheeled-Legged Robots Performing Mobile Manipulation Tasks
You, Y., Cheong, S., Chen, T. P., Chen, Y., Zhang, K., Acar, C., Lai, F. L., Adiwahono, A. H., & Tee, K. P. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9560948 · Cited by 14
“Learning-based methods are another popular solution to realize visual localization which either predicts matches for pose estimation [25] or directly regresses the pose [26].”
BackgroundOverview of 3D Scene Viewpoints evaluation method
Yan, Z., & Fei, G. (2019) · Virtual Reality & Intelligent Hardware · doi:10.1016/j.vrih.2019.01.001 · Cited by 14
Multimodal Localization for Embedded Systems: A Survey
Salhi, I., Porêba, M., Piriou, E., Gouet-Brunet, V., & Ojail, M. (2019) · Elsevier eBooks · doi:10.1016/b978-0-12-817358-9.00014-7 · Cited by 13
Civera, J., & Lee, S. H. (2019) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-030-28603-3_6 · Cited by 12
“[99] uses regression forests to predict correspondences between an RGB-D frame and the map, an approach that has been refined in [13,37,113] among others.”
MethodPoint-Graph Neural Network Based Novel Visual Positioning System for Indoor Navigation
Jung, T.-W., Jeong, C.-S., Kwon, S., & Jung, K.-D. (2021) · Applied Sciences · doi:10.3390/app11199187 · Cited by 12
A High-Accuracy Indoor-Positioning Method with Automated RGB-D Image Database Construction
Wang, R., Wan, W., Di, K., Chen, R., & Feng, X. (2019) · Remote Sensing · doi:10.3390/rs11212572 · Cited by 12
“In addition to the traditional visual-positioning method based on strict geometric relations, with the rapid development of deep learning in recent years scholars have proposed many learning based visual-positioning methods [39–41].”
MethodLazy Visual Localization via Motion AveragingPreprint
Dong, S., Liu, S., Guo, H., Chen, B., & Pollefeys, M. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2307.09981 · Cited by 12
“With the help of depth images, the series of works [10–12, 65, 74] achieve more accurate camera poses than SfM-based methods.”
MethodA Guide to Structureless Visual Localization
Panek, V., Zhou, Q., Ding, Y., Agostinho, S., Kúkelová, Z., Sattler, T., & Leal-Taixé, L. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-026-02780-9 · Cited by 12
“Rather than relying on descriptor matching to establish 2D-3D correspondences, scene coordinate regression methods train machine learning models to directly regress the corresponding 3D point position for a given input patch [19, 21, 20, 66, 41, 118, 109, 27, 17, 26, 131, 24, 76, 46, 125].”
Visual Robot Relocalization Based on Multi-Task CNN and Image-Similarity Strategy
Xie, T., Wang, K., Li, R., & Tang, X. (2020) · Sensors · doi:10.3390/s20236943 · Cited by 11
“[9,16] has been successfully applied to camera pose estimation, with taking RGB-D images as input, using depth images to create labels of scene coordinate and mapping each pixel from the camera coordinate system to the global scene coordinate system.”
MethodA Large-Scale Virtual Dataset and Egocentric Localization for Disaster Responses
Jeon, H., Im, S., Lee, B.-U., Rameau, F., Choi, D., Oh, J., Kweon, I. S., & Hebert, M. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3094531 · Cited by 11
“1c. Conventional visual localizations [12], [13], [14], [15] establish scene coordinates [12] that map image patches to corresponding dense 3D points in a scene model.”
BackgroundTowards Accurate Active Camera Localization
Fang, Q., Yin, Y., Fan, Q., Xia, F., Dong, S., Wang, S., Wang, J., Guibas, L. J., & Chen, B. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20080-9_8 · Cited by 10
“The popular scene coordinate regression approaches are implemented as a decision tree [9, 10, 27, 37, 42] or a convolution neural network [3, 4, 45].”
BackgroundMethodInfluential3D Image-based Indoor Localization Joint With WiFi Positioning
Lu, G., & Song, J. (2018) · International Conference on Multimedia Retrieval · doi:10.1145/3206025.3206070 · Cited by 11
“Kinect was also applied to localize the RGB-D image through regression forests [37, 42], which is limited to the usage in small rooms.”
MethodLaLaLoc: Latent Layout Localisation in Dynamic, Unvisited EnvironmentsPreprint
Howard-Jenkins, H., Ruiz-Sarmiento, J.-R., & Prisacariu, V. A. (2021) · Oxford University Research Archive (ORA) (University of Oxford) · doi:10.48550/arxiv.2104.09169 · Cited by 10
“Instead of regressing camera pose, scenecoordinate regression methods [29, 5, 6, 7, 34] densely predict 2D-3D correspondences between the query and the environment, which then allows for solving for pose via PnP.”
MethodReprojection Errors as Prompts for Efficient Scene Coordinate Regression
Liu, T.-R., Yang, H.-K., Liu, J.-M., Huang, C., Chiang, T.-C., Kong, Q., Kobori, N., & Lee, C. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73404-5_17 · Cited by 9
Camera Pose Estimation and Localization with Active Audio Sensing
Yang, K., Firman, M., Brachmann, E., & Godard, C. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19836-6_16 · Cited by 9
“Scene coordinate regression dispenses with the need for discrete feature matching by regressing image-to-scene correspondences directly, via random forests [82, 97, 8] or neural networks [7, 9, 11, 12].”
BackgroundAccurate Sparse Feature Regression Forest Learning for Real-Time Camera Relocalization
Duong, N.-D., Kacete, A., Soladié, C., Richard, P.-Y., & Royan, J. (2018) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2018.00079 · Cited by 9
“The first hybrid methods [26, 12, 29] are limited by the use of RGB-D sensors in the testing phase.”
MethodInfluentialDuong, N.-D., Soladié, C., Kacete, A., Richard, P.-Y., & Royan, J. (2019) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2019.102850 · Cited by 8
ROAM: A Rich Object Appearance Model with Application to Rotoscoping
Perez-Rua, J.-M., Mikšík, O., Crivelli, T., Bouthémy, P., Torr, P. H. S., & Pérez, P. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2904963 · Cited by 7
Self-citationDuong, N.-D., Kacete, A., Sodalie, C., Richard, P.-Y., & Royan, J. (2018) · IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) · doi:10.1109/ismar-adjunct.2018.00080 · Cited by 7
“The first hybrid methods [11, 24, 27] are limited by the use of RGB-D sensors in the testing phase.”
Method3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments
Ress, V., Meyer, J., Zhang, W., Skuddis, D., Soergel, U., & Norbert Haala, A. (2025) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xlviii-g-2025-1283-2025 · Cited by 6
“While early approaches used regression trees to perform this task (Shotton et al., 2013) (Valentin et al., 2015), neural networks have become established in modern SCR approaches (Li et al., 2018) (Dong et al., 2022) (Wang et al., 2024).”
Real-Time Direct Monocular SLAM With Learning-Based Confidence Estimation
Zhang, W., Yan, Z., Xiao, G., Feng, A., & Zuo, W. (2019) · IEEE Access · doi:10.1109/access.2019.2928578 · Cited by 6
“Confidence estimation based on learning methods have been adopted in stereo matching [21]–[24] and camera relocalization [25].”
MethodLive Collaborative Large-Scale Dense 3D Reconstruction Using Consumer-Grade Hardware
Golodetz, S., Cavallari, T., Lord, N. A., Prisacariu, V. A., Murray, D. W., & Torr, P. H. S. (2018) · IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) · doi:10.1109/ismar-adjunct.2018.00122 · Cited by 6
Self-citationPerspective from a Higher Dimension: Can 3D Geometric Priors Help Visual Floorplan Localization?
Chen, B., Kang, J., Yang, H., Zhong, P., & Wang, J. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3754504 · Cited by 5
“Unlike these classical methods, data-driven models utilize neural networks to regress the 3D coordinates of the pixels in the query image [6, 47] or directly regress the 6D camera pose of the input image [50, 53].”
Learning Shape-based Representation for Visual Localization in Extremely Changing Conditions
Jeon, H., Im, S., Oh, J., & Hebert, M. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9196842 · Cited by 5
“Previous works [1], [2], [3], [4] on visual localization have shown promising performance by establishing scene coordinates [1] that map image patches to corresponding dense 3D points in a scene model.”
BackgroundEuler Angles Based Loss Function for Camera Localization with Deep Learning
Fang, Q., Zhao, K., Tang, D., Zhou, Z., Zhou, Y., Hu, T., & Zhou, H. (2018) · IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems · doi:10.1109/cyber.2018.8688359 · Cited by 5
Fast and Lightweight Scene Regressor for Camera RelocalizationPreprint
Bui, T. B., Tran, D.-T., & Lee, J. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2212.01830 · Cited by 5
“Several follow-up studies have improved this approach in terms of accuracy [28] and learning camera localization on-the-fly [29].”
BackgroundResearch on an Optimal Path Planning Method Based on A* Algorithm for Multi-View Recognition
Li, X., He, Q., Yang, Q., Wang, N., Wu, H., & Yang, X. (2022) · Algorithms · doi:10.3390/a15050171 · Cited by 4
“Valentin et al. [27] studied the uncertainty in forest regression to achieve accurate camera positioning.”
BackgroundThe Relocalization of SLAM Tracking Based on Spherical Cameras
Chang, Q., Liu, Q., Xin, Y., Huang, Y., Ren, F., & Cui, Y. (2021) · IEEE Access · doi:10.1109/access.2021.3130928 · Cited by 4
Image-similarity-based Convolutional Neural Network for Robot Visual Relocalization
Wang, L., Li, R., Sun, J., Seah, H. S., Quah, C. K., Zhao, L., & Tandianus, B. (2020) · Sensors and Materials · doi:10.18494/sam.2020.2549 · Cited by 4
Learning-based Camera Relocalization with Domain Adaptation via Image-to-Image Translation
Yang, C., Liu, Y., & Zell, A. (2021) · International Conference on Unmanned Aircraft Systems (ICUAS) · doi:10.1109/icuas51884.2021.9476673 · Cited by 4
Purely Geometric Scene Association and Retrieval - A Case for Macro Scale 3D GeometryPreprint
Sawhney, R., Li, F., Christensen, H. I., & Isbell, C. L. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.01343 · Cited by 4
“Method Spr[40] [5]C DSc[4] [40] [16] [46] [5] D[40] R VDR”
On Localizing a Camera from a Single ImagePreprint
Ghosh, P., Liu, X., Qiu, H., Vieira, M. A. M., Sukhatme, G. S., & Govindan, R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.10664 · Cited by 4
“However, such techniques suffer from two inherent problems: (1) the location of the image is not same as the location of the camera, and (2) the database of images is often sparse and might not contain any matching images [37].”
BackgroundLangerman, J., Qiu, Z., Sörös, G., Sebok, D., Wang, Y., & Huang, H. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.14741 · Cited by 4
“One family of regression approaches use learned models [34] [40] [3] [7] to perform SCR and then input point samples from the intermediate scene coordinate map to a PnP-RANSAC pose estimator.”
MethodNaVIP: An Image-Centric Indoor Navigation Solution for Visually Impaired PeoplePreprint
Yu, J., Zhang, Y., Aila, B., & Namboodiri, V. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2410.18109 · Cited by 4
“Direct localization can instantly relocalize the camera pose from the query image [88, 89, 90, 30, 91, 92], thus requiring a smaller memory footprint compared to indirect methods.”
Huang, Z., Yu, H., Shentu, Y., Yuan, J., & Zhang, G. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2503.19358 · Cited by 4
“SCR meth-ods do not directly predict camera pose but instead infer the dense scene coordinates [4, 15, 42, 47, 56, 58, 61].”
Implicit neural representation model for camera relocalization in multiple scenes
Yao, S., Cheng, Y., Yang, F., & Mozerov, M. G. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.111791 · Cited by 3
Unsupervised Simultaneous Learning for Camera Re-Localization and Depth Estimation from Video
Taguchi, S., & Hirose, N. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9982213 · Cited by 3
“Originally, scene coordinate regression was proposed for RGB-D-based re-localization in indoor environments [4], [18]–[20].”
BackgroundLeveraging Spatial Attention and Edge Context for Optimized Feature Selection in Visual Localization
Istighfarin, N. F., & Jo, H. (2025) · International Journal of Control Automation and Systems · doi:10.1007/s12555-024-0487-y · Cited by 3
“The initial proposed SCR [31,32] required RGB-D input, making it a drawback when we use inexpensive sensors only, i.e. RGB camera.”
RnR: Retrieval and Reprojection Learning Model for Camera Localization
Yang, S., & Shi, D. (2021) · IEEE Access · doi:10.1109/access.2021.3061634 · Cited by 3
Geometrical Features based Visual Relocalization for Indoor Service Robot
Wang, J., Dai, D., Wang, P., Zhao, H., Xu, M., & Chen, Z. (2020) · International Conference on Communication, Image and Signal Processing (CCISP) · doi:10.1109/ccisp51026.2020.9273485 · Cited by 2
“Then, various machine learning based methods, mainly random forests based [5]–[7] and deep learning based methods [8]–[12] have been proposed to improve camera relocalization system.”
MethodFast and Lightweight Scene Regressor for Camera Relocalization
Bui, B.-T., Tran, D.-T., & Lee, J. (2024) · IEEE/SICE International Symposium on System Integration (SII) · doi:10.1109/sii58957.2024.10417635 · Cited by 2
“Several follow-up studies have improved this approach in terms of accuracy [28] and learning camera localization on-the-fly [29].”
BackgroundChai, X., Yang, Z., Tan, X., Zhu, M., Zhong, C., & Shi, J. (2025) · Sensors · doi:10.3390/s25061920 · Cited by 2
“In Section 4.2, we present the performance of our improved model compared to the baseline ACE model on two indoor datasets, 7scenes [32] and 12scenes [33], as well as the outdoor dataset wayspots.”
InfluentialWu, J., Ma, L., & Hu, X. (2016) · International Conference on Intelligent Control and Information Processing · doi:10.1109/icicip.2016.7885894 · Cited by 2
Image-Based Localization Using Deep Neural Networks
Li, X. (2017) · Aaltodoc (Aalto University) · Cited by 2
Barroso-Laguna, A., Cavallari, T., Prisacariu, V. A., & Brachmann, E. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2510.00978 · Cited by 2
“Traditionally, SCR relied on random forest (Shotton et al., 2013; Valentin et al., 2015; Brachmann et al., 2016; Cavallari et al., 2017; Cavallari et al., 2019), but in recent years, SCR improved their accuracy by employing convolutional neural networks (Brachmann & Rother, 2021; Cavallari et al.,…”
Incorporating Global Correlation and Local Aggregation for Efficient Visual Localization
Xie, D., Lu, J., Song, Z., & Chen, X. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-46308-2_6 · Cited by 1
Robot Navigation and Obstacle Avoidance
Yang, X., Yin, B., & Wei, X. (2025) · Springer eBooks · doi:10.1007/978-981-96-3177-3_4 · Cited by 1
Monocular Camera Localization in Known Environments: An In-Depth Review
Yan, H., Lau, A., & Fan, H. (2026) · Applied Sciences · doi:10.3390/app16052332 · Cited by 1
Robust Scene Coordinate Regression via Geometrically-Consistent Global Descriptors
Nguyen, S. T., Fontan, A., Milford, M. J., & Fischer, T. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00792 · Cited by 1
“Scene coordinate regression (SCR) methods [7,9,41,55, 59,60] have emerged as an attractive alternative, excelling in small-scale environments in both accuracy and memory efficiency.”
Back to the Context: Tuning Visual Localization Using Structural and Edge Context Within Image
Istighfarin, N. F., Lee, S., & Jo, H. (2024) · IEEE Access · doi:10.1109/access.2024.3483963 · Cited by 1
“The first introduced SCR [8], [27] proposes SCR using RGB-D input as their input.”
Methodcvpaper.challenge in 2015 - A review of CVPR2015 and DeepSurveyPreprint
Kataoka, H., Miyashita, Y., Yamabe, T. K., Shirakabe, S., Sato, S., Hoshino, H., Kato, R., Abe, K., Imanari, T., Kobayashi, N., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.08247 · Cited by 1
cvpaper.challenge in 2016: Futuristic Computer Vision through 1,600 Papers SurveyPreprint
Kataoka, H., Shirakabe, S., He, Y., Ueta, S., Suzuki, T., Abe, K., Kanezaki, A., Morita, S., Yabe, T., Kanehara, Y., et al. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1707.06436 · Cited by 1
“[554], [555], [556], [557], [558], [559], [560], [561], [562], [563],”
Active Visual Localization in Partially Calibrated Environments.Preprint
Yin, Y., Fan, Q., Xia, F., Fang, Q., Dong, S., Guibas, L. J., & Chen, B. (2020) · arXiv (Cornell University) · Cited by 1
PoI: A Filter to Extract Pixel of Interest from Novel Views for Scene Coordinate RegressionPreprint
Li, F., Song, Q., Zhang, C., Shuai, H., & Huang, R. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2502.04843 · Cited by 1
Deep Learning Meets Visual Localization
Zhou, Q., Leal-Taixé, L., & München, T. U. (2022) · mediaTUM (Technical University of Munich) · Cited by 1
Hybrid Machine Learning and Geometric Approaches for Single RGB Camera Relocalization
Duong, N.-D. (2019) · theses.fr (ABES) · doi:10.70675/8e4ecee4z215fz435az93aez8cb46cfd16fa · Cited by 1
Multi-descriptor retrieval in digitalized photographs collections
Bhowmik, N. (2017) · theses.fr (ABES) · doi:10.70675/1a62ff5dz0e8fz4d53zb0cczc01b30ac200d
Euler angles based loss function for camera relocalization with Deep learningPreprint
Fang, Q., & Hu, T. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1802.08851
GCP-SLAM: LSD-SLAM with Learning-Based Confidence Estimation
Feng, A., Zhang, W., Yan, Z., & Zuo, W. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-75786-5_22
Efficient algorithms and systems for multi-agent mapping and world-scale real-time localization
Lynen, S. (2019) · Repository for Publications and Research Data (ETH Zurich) · doi:10.3929/ethz-b-000350558
Decoupling Features and Coordinates for Few-shot RGB RelocalizationPreprint
Dong, S., Wu, S., Zhuang, Y., Xu, K., Zhang, S., & Chen, B. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1911.11534
“There are an excellent series of works on regressing 3D coordinates of image pixels using learned random forests [27, 29, 5, 4], reaching the state-of-the-art performance.”
BackgroundDeep Learning Methods for Semantic Matching, Image Retrieval and Camera Relocalization
Laskar, Z. (2020) · Aaltodoc (Aalto University)
Model-Based Self-Supervision for Fine-Grained Image Understanding
Schmidt, T. (2019) · ResearchWorks at the University of Washington (University of Washington)
A Miniature Vision-Based Localization System for Indoor BlimpsPreprint
Ma, S. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2408.06648
Istighfarin, N. F., Choi, B., & Jo, H. (2025) · International Journal of Control Automation and Systems · doi:10.1007/s12555-025-0527-2
“Despite these advantages, both early SCR meth-ods—such as random forest-based or fully neural approaches [8,22,23]—suffer from long training times, limiting their scalability.”
SANet: Scene agnostic network for camera localization
Bai, Z., & Bai, Z. (2021) · Summit (Simon Fraser University)
Deep Learning Methods for Point Matching, Visual Localization and 3D Reconstruction
Wang, S. (2024) · Aaltodoc (Aalto University)
A Robot Visual Localization Architecture Using Hybrid Map with Explicit-Implicit Combination
Guan, P., Gong, X., Li, J., Cao, Z., & Yu, J. S. (2025) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio66223.2025.11376024
“Representative methods estimate per-pixel 3D world coordinates using random forests [22] or scene coordinate regression networks [9] and compute camera pose via RANSAC-based PnP formulations.”
CoCo-SCR: Covisibility-Centric Scene Coordinate Regression for Large-Scale UAV Visual Localization
Zheng, L., Lin, W., Ren, K., & Chen, Q. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3708150
“Early SCR approaches mainly relied on random forests [54], [55], [56], [57], [58].”
Xin’an, Q. I. U., Wang, L., Dong, Z., Xiao, X., Liu, Z., Zhu, L., Wang, J., & Xu, H. (2026) · Sensors · doi:10.3390/s26165070
“Scene coordinate regression (SCR) predicts a 3D coordinate for each pixel and then applies PnP. Early random-forest regressors [37–39] were followed by differentiable DSAC [40], DSAC++ [2], hierarchical prediction [41], and DSAC* [3].”
Sun, S., & Luo, J. (2026) · Research Square · doi:10.21203/rs.3.rs-8936961/v1
AIRLoc: An IoU-Guided Adaptive Fusion Method for Floorplan-Based Visual Localization
Peng, S., Wu, X., Yu, H., Ou, Y. X., Liao, Z., Wang, S., & Zhong, Y. (2026) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2026.3741770
“Scene coordinate regression methods predict dense 3D correspondences [23], [24], [27], while pose regression methods directly regress the 6-DoF pose [18], [19], [21].”
A survey on heterogeneous information fusion for robust visual re-localization
Kuang, Y., Ouyang, M., 李起伟, Yang, Y., Ding, W., An, X., & Zhang, X. (2026) · Information Fusion · doi:10.1016/j.inffus.2026.104854
Liu, E., Hu, Z., Lei, X., Duan, W., Qi, Y., Liu, M., Wang, J., & Gan, T. (2026) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3767308.3835245
Metadata
- Authors
- Valentin, J., Niebner, M., Shotton, J., Fitzgibbon, A., Izadi, S., & Torr, P.
- Affiliations
- University of Oxford (GB)
- Stanford University (US)
- Microsoft (United States) (US)
- Venue
- 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2015-06
- Pages
- 4400-4408
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W1893935112
- Cited by
- 156
- References
- 23
- Citations by year
- 2026: 4 · 2025: 9 · 2024: 13 · 2023: 7 · 2022: 13 · 2021: 19 · 2020: 19 · 2019: 25 · 2018: 20 · 2017: 15 · 2016: 12
- FWCI
- 27.16
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- Funders
- Leverhulme Trust
- Engineering and Physical Sciences Research Council (EP/I001107/2)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Keywords
- camera relocalization, regression forests, uncertainty estimation, camera pose optimization, visual SLAM
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.