Are Large-Scale 3D Models Really Necessary for Accurate Visual Localization?
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Accurate visual localization is a key technology for autonomous navigation. 3D structure-based methods employ 3D models of the scene to estimate the full 6 degree-of-freedom (DOF) pose of a camera very accurately. However, constructing (and extending) large-scale 3D models is still a significant challenge. In contrast, 2D image retrieval-based methods only require a database of geo-tagged images, which is trivial to construct and to maintain. They are often considered inaccurate since they only approximate the positions of the cameras. Yet, the exact camera pose can theoretically be recovered when enough relevant database images are retrieved. In this paper, we demonstrate experimentally that large-scale 3D models are not strictly necessary for accurate visual localization. We create reference poses for a large and challenging urban dataset. Using these poses, we show that combining image-based methods with local reconstructions results in a higher pose accuracy compared to state-of-the-art structure-based methods, albeight at higher run-time costs. We show that some of these run-time costs can be alleviated by exploiting known database image poses. Our results suggest that we might want to reconsider the need for large-scale 3D models in favor of more local models, but also that further research is necessary to accelerate the local reconstruction process.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Distinctive Image Features from Scale-Invariant Keypoints
Lowe, D. G. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000029664.99615.94 · Cited by 55,589
Fischler, M. A., & Bolles, R. C. (1981) · Communications of the ACM · doi:10.1145/358669.358692 · Cited by 25,893
Structure-from-Motion Revisited
Schonberger, J. L., & Frahm, J.-M. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.445 · Cited by 6,433
Video Google: a text retrieval approach to object matching in videos
Sivic, & Zisserman (2003) · Proceedings Ninth IEEE International Conference on Computer Vision · doi:10.1109/iccv.2003.1238663 · Cited by 6,367 · cited 1×
“They employ standard retrieval techniques such as Bag-of-Words (BoW) representations with inverted files [35], followed by fast spatial verification [36], [37], or more compact representations such as VLAD or Fischer Vectors [38], [39].”
MethodProduct Quantization for Nearest Neighbor Search
Jégou, H., Douze, M., & Schmid, C. (2011) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2010.57 · Cited by 2,956 · cited 1×
“Storing 4096-dimensional VLAD [16], [42] for all 1,062,468 database images requires 17.4 GB, and the requirements can be reduced further, e.g., using product quantization [86], with negligible loss in performance [16].”
MethodInfluentialPoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization
Kendall, A., Grimes, M., & Cipolla, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.336 · Cited by 2,489
Hamming Embedding and Weak Geometric Consistency for Large Scale Image Search
Jegou, H., Douze, M., & Schmid, C. (2008) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-88682-2_24 · Cited by 1,901
Aggregating Local Image Descriptors into Compact Codes
Jegou, H., Perronnin, F., Douze, M., Sanchez, J., Perez, P., & Schmid, C. (2012) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2011.235 · Cited by 1,503
Visual Place Recognition: A Survey
Lowry, S., Sünderhauf, N., Newman, P., Leonard, J. J., Cox, D., Corke, P., & Milford, M. J. (2016) · IEEE Transactions on Robotics · doi:10.1109/tro.2015.2496823 · Cited by 1,079 · cited 1×
“D ETERMINING the location from which a photo was taken is a key challenge in the navigation of autonomous vehicles such as self-driving cars and drones [1], robotics [2], mobile Augmented Reality [3], [4], and Structure-from-Motion (SfM) [5], [6], [7], [8].”
BackgroundGeometric Loss Functions for Camera Pose Regression with Deep Learning
Kendall, A., & Cipolla, R. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.694 · Cited by 803 · cited 3×
“Camera pose regression techniques such as Pose-Net and its variants [64], [65], [66], [67] replace the full localization stack by a single convolutional neural network (CNN) that is trained to directly predict a 6DOF camera pose.”
BackgroundMethodResultBenchmarking 6DOF Outdoor Visual Localization in Changing Conditions
Sattler, T., Maddern, W., Toft, C., Torii, A., Hammarstrand, L., Stenborg, E., Safari, D., Okutomi, M., Pollefeys, M., Sivic, J., et al. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00897 · Cited by 699 · cited 5×
“In addition, training them on larger datasets still seems to be an open problem [68], [69].”
BackgroundMethodResultInfluentialSelf-citationImage-Based Localization Using LSTMs for Structured Feature Correlation
Walch, F., Hazirbas, C., Leal-Taixe, L., Sattler, T., Hilsenbeck, S., & Cremers, D. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.75 · Cited by 521 · cited 3×
“Camera pose regression techniques such as Pose-Net and its variants [64], [65], [66], [67] replace the full localization stack by a single convolutional neural network (CNN) that is trained to directly predict a 6DOF camera pose.”
BackgroundMethodSelf-citationLearning Less is More - 6D Camera Localization via 3D Surface Regression
Brachmann, E., & Rother, C. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00489 · Cited by 399
Get Out of My Lab: Large-scale, Real-Time Visual-Inertial Localization
Lynen, S., Sattler, T., Bosse, M., Hesch, J., Pollefeys, M., & Siegwart, R. (2015) · Robotics: Science and Systems XI · doi:10.15607/rss.2015.xi.037 · Cited by 243 · cited 2×
“D ETERMINING the location from which a photo was taken is a key challenge in the navigation of autonomous vehicles such as self-driving cars and drones [1], robotics [2], mobile Augmented Reality [3], [4], and Structure-from-Motion (SfM) [5], [6], [7], [8].”
BackgroundSelf-citationExploiting uncertainty in regression forests for accurate camera relocalization
Valentin, J., Niebner, M., Shotton, J., Fitzgibbon, A., Izadi, S., & Torr, P. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7299069 · Cited by 156 · cited 1×
“Rather than replacing the complete localization pipeline, scene coordinate regression techniques [72], [73], [74], [75], [76], [77] only replace the 2D-3D matching stage while keeping the RANSAC-based pose estimation stage.”
Method
Cited by
D2-Net: A Trainable CNN for Joint Description and Detection of Local Features
Dusmanu, M., Rocco, I., Pajdla, T., Pollefeys, M., Sivic, J., Torii, A., & Sattler, T. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00828 · Cited by 1,179
Self-citationThe ApolloScape Dataset for Autonomous Driving
Huang, X., Cheng, X., Geng, Q., Cao, B., Zhou, D., Wang, P., Lin, Y., & Yang, R. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2018.00141 · Cited by 679
Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions
Sattler, T., Maddern, W., Toft, C., Torii, A., Hammarstrand, L., Stenborg, E., Safari, D., Okutomi, M., Pollefeys, M., Sivic, J., et al. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00897 · Cited by 699
Self-citationThe ApolloScape Open Dataset for Autonomous Driving and Its Application
Huang, X., Wang, P., Cheng, X., Zhou, D., Geng, Q., & Yang, R. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2926463 · Cited by 615
The ApolloScape Open Dataset for Autonomous Driving and its Application
Huang, X., Wang, P., Cheng, X., Zhou, D., Geng, Q., & Yang, R. (2018) · arXiv · doi:10.48550/arxiv.1803.06184 · Cited by 615
Metadata
- Authors
- Torii, A., Taira, H., Sivic, J., Pollefeys, M., Okutomi, M., Pajdla, T., & Sattler, T.
- Venue
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Published
- 2021-03-01
- Volume
- 43
- Issue
- 3
- Pages
- 814-829
- ISSN
- 0162-8828 (ISSN-L)
- 2160-9292
- 1939-3539
- OpenAlex
- W2973689592
- Cited by
- 126
- References
- 15
- FWCI
- 15.74
- Type
- journal article
- Language
- en
- Open access
- green
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.