SURF: Speeded Up Robust Features
Open accessThe DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index; none of the 18 sources its claims cite is checked for support yet.
Overview
CoChat Summary
TL;DR
SURF is a scale- and rotation-invariant interest point detector and descriptor that uses integral images, Hessian-based detection, and Haar-wavelet responses to be faster than prior methods while matching or exceeding their repeatability and distinctiveness. It reports strong results on standard benchmarks and a museum object recognition task.
Key findings
- The Fast-Hessian detector is more than 3× faster than DoG and 5× faster than Hessian-Laplace, with comparable or better repeatability on Graffiti, Leuven, Boat, and Wall; on the first Graffiti image (800×640), computation times were 120 ms for Fast-Hessian, 400 ms for DoG, 650 ms for Hessian-Laplace, and 1800 ms for Harris-Laplace.
- SURF’s 64-dimensional descriptor outperformed GLOH, SIFT, and PCA-SIFT in most comparisons, sometimes by more than 10% improvement in recall for the same precision; SURF-128 was slightly better but slower to match.
- Joint detector-descriptor computation times on the first Graffiti image were 255 ms for U-SURF, 354 ms for SURF, 391 ms for SURF-128, and 1036 ms for SIFT.
- In museum object recognition, recognition rates were 85.7% for SURF-128, 83.8% for U-SURF, 82.6% for SURF, 78.3% for GLOH, 78.1% for SIFT, and 72.3% for PCA-SIFT.
Methods
- Design: a scale- and rotation-invariant detector/descriptor. The detector uses the determinant of the Hessian with box-filter approximations evaluated via integral images; the descriptor uses Haar-wavelet responses in a 4×4 subregion grid, producing a 64-dimensional vector, with an upright U-SURF variant and an extended SURF-128 variant.
- Data: standard evaluation sequences from Mikolajczyk (including Graffiti, Wall, Boat, Leuven, Bikes, Trees, and Ubc; Bark excluded from descriptor evaluations) and a real-life museum object recognition database of 216 images of 22 objects, with 116 test images at 320×240.
- Analysis: detector repeatability scores, descriptor recall-(1-precision) graphs under similarity-threshold and nearest-neighbour-ratio matching, recognition rates, and computation timings on a standard Linux PC (Pentium IV, 3 GHz).
Limitations
- The paper has no explicit limitations section. It states that space limitations prevent showing results on all sequences, and it assumes skew, anisotropic scaling, and perspective effects are second-order effects covered only by overall descriptor robustness.
- Evident from the text: the method is scale- and rotation-invariant but not affine-invariant; Fig. 4 notes the interest points are not affine invariant, and the descriptor does not use colour.
- Evident from the text: the object recognition experiment uses small 320×240 test images and matching without additional geometric constraints, which the authors say would reduce false positives but were omitted for comparison purposes.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures








Verification
Claims the paper makes
0 Valid · 2 Not valid · 12 Not checked · 30 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The number of interest points found is on average very similar for all detectors.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The 4 × 4 sub-region division solution provided the best results.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“SURF outperformed the other descriptors for almost all the comparisons.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The SURF descriptor outperforms the other descriptors in a systematic and significant way, with sometimes more than 10% improvement in recall for the same level of precision.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As shown in the results section, the performance is comparable to the one using the discretised and cropped Gaussians.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
8 more claims
Create a free account to see all 44 claims and every source verdict.
- It may state two separate claims as one. 24
- Probably not a substantial claim. 13
- The quote only partly states this claim. 1
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
“This transition marked a new era, where we moved beyond the classic approaches and task-specific feature engineering of earlier days [Lowe 2004; Bay et al. 2006; Rosten and Drummond 2006] towards models that could be trained once over large amounts of data, and then adapted for a broad variety of tasks, such as image recognition, object detection, and image segmentation [Krizhevsky et al. 2012; Sz”
From On the Opportunities and Risks of Foundation Models (2021) · cites this paper as Bay et al. 2006
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
Rapid object detection using a boosted cascade of simple features
Viola, P., & Jones, M. (2005) · Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001 · doi:10.1109/cvpr.2001.990517 · Cited by 18,300
Object recognition from local scale-invariant features
Lowe, D. G. (1999) · Proceedings of the Seventh IEEE International Conference on Computer Vision · doi:10.1109/iccv.1999.790410 · Cited by 16,472
A Combined Corner and Edge Detector
Harris, C., & Stephens, M. (1988) · Procedings of the Alvey Vision Conference 1988 · doi:10.5244/c.2.23 · Cited by 12,901
A performance evaluation of local descriptors
Mikolajczyk, K., & Schmid, C. (2005) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2005.188 · Cited by 6,723 · cited 14×
“This PCA-SIFT yields a 36-dimensional descriptor which is fast for matching, but proved to be less distinctive than SIFT in a second comparative study by Mikolajczyk et al. [8] and slower feature computation reduces the effect of fast matching.”
BackgroundMethodResultInfluentialMarr, D., & Hildreth, E. (1980) · Proceedings of the Royal Society of London. Series B. Biological Sciences · doi:10.1098/rspb.1980.0020 · Cited by 6,510
PCA-SIFT: a more distinctive representation for local image descriptors
Ke, Y., & Sukthankar, R. (2004) · Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004. · doi:10.1109/cvpr.2004.183 · Cited by 3,932
Robust wide-baseline stereo from maximally stable extremal regions
Matas, J., Chum, O., Urban, M., & Pajdla, T. (2004) · Image and Vision Computing · doi:10.1016/j.imavis.2004.02.006 · Cited by 3,686
Scalable Recognition with a Vocabulary Tree
Nister, D., & Stewenius, H. (2006) · 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2 (CVPR'06) · doi:10.1109/cvpr.2006.264 · Cited by 3,579
Scale & Affine Invariant Interest Point Detectors
Mikolajczyk, K., & Schmid, C. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000027790.02288.f2 · Cited by 3,436
Cited by
Items 7,601–7,700 of 14,752
Parallel Image Stitching Based on Multithreaded Processing on GPU
Nakov, O., Mihaylova, E., Lazarova, M. K., et al. (2018) · International Conference on Intelligent and Innovative Computing Applications · doi:10.1109/iconic.2018.8601253 · Cited by 5
Soft decision optimization method for robust fundamental matrix estimation
Xiao, C.-B., Feng, D., & Yuan, M. (2019) · Machine Vision and Applications · doi:10.1007/s00138-019-01019-7 · Cited by 5
“The most popular local invariant features include scale invariant feature transform (SIFT) [4] and speeded-up robust features (SURF) [5].”
An Improved Illumination Invariant SURF Image Feature Descriptor
Geng, Z., & Qiao, Y. Q. (2017) · International Conference on Virtual Reality and Visualization · doi:10.1109/icvrv.2017.00090 · Cited by 6
“The feature-based alignment algorithms is commonly used in image matching, and the SIFT algorithm([1]) is widely used because it’s robust in the change of scale, rotation and illumination, But the SIFT descriptors occupies large storage space, the SURF algorithm([2]) is a more successful improvement.”
MethodReal-time Human Detection and Tracking in Infrared Video Feed
Fernando, H., Perera, I., & Silva, C. K. D. (2019) · Moratuwa Engineering Research Conference (MERCon) · doi:10.1109/mercon.2019.8818862 · Cited by 5
“SURF features [20] have been used by Hamdoun et al.”
MethodInfluentialDirect-ORB-SLAM: Direct Monocular ORB-SLAM
Cui, L.-Y., Ma, C.-W., & Wen, F. (2019) · Journal of Physics, Conference Series · doi:10.1088/1742-6596/1345/3/032016 · Cited by 5
Manikandan, S., Nigam, C., Ramakrishnan, S., et al. (2020) · IEEE India Geoscience and Remote Sensing Symposium (InGARSS) · doi:10.1109/ingarss48198.2020.9358941 · Cited by 5
A New Visual Comfort-Based Stereoscopic Image Retargeting Method
Cho, S. H., & Kang, H.-B. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-41914-0_29 · Cited by 5
EXTERIOR ORIENTATION ESTIMATION OF OBLIQUE AERIAL IMAGERY USING VANISHING POINTS
Verykokou, S., & Ioannidis, C. (2016) · 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/isprsarchives-xli-b3-123-2016 · Cited by 5
Girdhar, Y., & Dudek, G. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5980094 · Cited by 6
A Novel SIFT-Like-Based Approach for FIR-VS Images Registration
Aguilera, C., Barrera, F., Sappa, A., et al. (2012) · International Conference on Quantitative InfraRed Thermography (QIRT) · doi:10.21611/qirt.2012.149 · Cited by 5
Online learning for automatic segmentation of 3D data
Tombari, F., Stefano, L. D., & Giardino, S. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6048294 · Cited by 7
Topological Mapping and Localization Using Omnidirectional Vision.
Valgren, C. (2007) · Cited by 5
Challenging the Security of CBIR Systems
Kijak, E., Furon, T., & Amsaleg, L. (2009) · HAL · Cited by 5
HARMONIC DEFORMATION MODEL FOR EDGE BASED TEMPLATE MATCHING
Hofhauser, A., Steger, C. T., & Navab, N. (2008) · International Conference on Computer Vision Theory and Applications · doi:10.5220/0001071800750082 · Cited by 5
Re-identification of Visual Targets in Camera Networks: A Comparison of Techniques
Figueira, D., & Bernardino, A. J. M. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-21593-3_30 · Cited by 5
Knuckle Biometrics for Human Identification
Choraś, M., & Kozik, R. (2010) · Advances in intelligent and soft computing · doi:10.1007/978-3-642-16295-4_10 · Cited by 5
Feature Detection Applied to Context-Aware Blind Guidance Support
Fernandes, H., Sousa, A., Paredes, H., Filipe, V. M. D. J., & Barroso, J. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-20687-5_13 · Cited by 5
“SURF cannot be used for commercial purposes without permission of the patent holder [11].”
BackgroundInvariant backpropagation: how to train a transformation-invariant neural networkPreprint
Demyanov, S., Bailey, J., Ramamohanarao, K., & Leckie, C. A. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1502.04434 · Cited by 12
“Two other popular tools are SIFT (Lowe, 2004) and SURF (Bay et al., 2006) features, which are invariant to translation, scale and rotation.”
MethodVasudevan, S. (2008) · Repository for Publications and Research Data (ETH Zurich) · doi:10.3929/ethz-a-005552714 · Cited by 6
Comparing Feature Detectors: A bias in the repeatability criteria, and how to correct itPreprint
Rey-Otero, I., Delbracio, M., & Morel, J. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1409.2465 · Cited by 5
Lyapunov-based nonlinear estimation methods with applications to machine vision
Dixon, W. E., & Dani, A. P. (2011) · PhDT · Cited by 5
Building reconstruction from terrestrial video image sequences
Tian, Y. (2011) · University of Twente Research Information · doi:10.3990/1.9789061643067 · Cited by 5
Covariance of Motion and Appearance Featuresfor Spatio Temporal Recognition TasksPreprint
Bhattacharya, S., Souly, N., & Shah, M. A. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1606.05355 · Cited by 5
Mapping and Localization in Urban Environments Using Cameras
Lategahn, H. (2013) · Repository KITopen (Karlsruhe Institute of Technology) · doi:10.5445/ksp/1000037227 · Cited by 8
Comparison of local descriptors for image registration of geometrically-complex 3D scenes
Cheng, D. C., Xie, S. Q., & Hämmerle, E. (2007) · International Conference on Mechatronics and Machine Vision in Practice (M2VIP) · doi:10.1109/mmvip.2007.4430732 · Cited by 5
“Experimental work has shown that the SURF descriptor has a significant advantage in terms of performance and computation time when compared with the SIFT descriptor [7].”
BackgroundMethodInfluentialSchaumberg, A. J., Sirintrapun, S. J., Al‐Ahmadie, H. A., et al. (2016) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/097246 · Cited by 5
Automated 3D vision-based tracking of construction entities
Park, M.-W. (2012) · SMARTech Repository (Georgia Institute of Technology) · Cited by 6
Image Retrieval using SURF Features
Kumar, A., & Batra, S. (2011) · Cited by 5
Depth maps estimation and use for 3DTV
Sourimant, G. (2010) · HAL (Le Centre pour la Communication Scientifique Directe) · Cited by 5
Happy Travelers Take Big Pictures: A Psychological Study with Machine Learning and Big DataPreprint
Liang, X., Fan, L., Loh, Y. P., Liu, Y., & Tong, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1709.07584 · Cited by 8
Multi-Image Semantic Matching by Mining Consistent FeaturesPreprint
Wang, Q., Zhou, X., & Daniilidis, K. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.07641 · Cited by 5
Visual recognition in art using machine learning
Crowley, E. J. (2017) · Oxford University Research Archive (ORA) (University of Oxford) · doi:10.5287/ora-py9q0enrx · Cited by 8
Online model-based multi-scale pose estimation
Kempter, T. (2011) · TUGraz OPEN Library (Graz University of Technology) · doi:10.3217/cdppt-8y148 · Cited by 5
A Quaternion-based Certifiably Optimal Solution to the Wahba Problem with OutliersPreprint
Yang, H., & Carlone, L. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.12536 · Cited by 5
Image Retrieval with Fisher Vectors of Binary FeaturesPreprint
Uchida, Y., Sakazawa, S., & Satoh, S. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1609.08291 · Cited by 5
“Binary features are one or two orders of magnitude faster than the Scale Invariant Feature Transform (SIFT) [27] or Speeded Up Robust Features (SURF) [8] features in detection and description, while providing comparable performance [39, 13].”
A Joint Optimization Approach of LiDAR-Camera Fusion for Accurate Dense 3D ReconstructionsPreprint
Zhen, W., Hu, Y., Liu, J., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.00930 · Cited by 5
Liu, Z., Suo, C., Zhou, S., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.13030 · Cited by 5
A Possible Reason for why Data-Driven Beats Theory-Driven Computer VisionPreprint
Tsotsos, J. K., Kotseruba, I., Andreopoulos, A., & Wu, Y. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1908.10933 · Cited by 5
Simultaneous Localization and Mapping ( SLAM )
Birk, A., & Pfingsthorn, M. F. (2016) · Wiley eBooks · doi:10.1002/047134608x.w8322 · Cited by 12
Efficient Neighbourhood Consensus Networks via Submanifold Sparse ConvolutionsPreprint
Rocco, I., Arandjelović, R., & Šivic, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.10566 · Cited by 19
Dense Semantic 3D Map Based Long-Term Visual Localization with Hybrid FeaturesPreprint
Shi, T., Cui, H., Song, Z., & Shen, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.10766 · Cited by 11
IV-SLAM: Introspective Vision for Simultaneous Localization and MappingPreprint
Rabiee, S., & Biswas, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.02760 · Cited by 19
“Selecting interest points on the images for extracting these descriptors is traditionally done by convolving the image with specific filters [11] or using first-order approximations ar X iv :2 00 8.”
Privacy-Preserving Image Retrieval Based on Additive Secret SharingPreprint
Xia, Z., Gu, Q., Xiong, L., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2009.06893 · Cited by 5
Plant leaf classification using dual path convolutional neural networks
Rizk, S. (2019) · Cited by 5
Localization of Autonomous Vehicles: Proof of Concept for A Computer Vision ApproachPreprint
Zahedian, S., Sadabadi, K. F., & Nohekhan, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.02785 · Cited by 6
DiscoBox: Weakly Supervised Instance Segmentation and Semantic Correspondence from Box SupervisionPreprint
Lan, S., Yu, Z., Choy, C., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2105.06464 · Cited by 5
Kimera-Multi: Robust, Distributed, Dense Metric-Semantic SLAM for Multi-Robot SystemsPreprint
Tian, Y., Chang, Y., Arias, F. H., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.14386 · Cited by 5
Hwang, S.-Y., Lee, J.-H., Ha, C., et al. (2023) · Preprints.org · doi:10.20944/preprints202312.0190.v1 · Cited by 5
Multi-Domain Indoor Dataset for Visual Place Recognition and Anomaly Detection by Mobile Robots
Woźniak, P., Krzeszowski, T., & Kwolek, B. (2025) · Scientific Data · doi:10.1038/s41597-025-05124-3 · Cited by 9
“Methods that determine hand-crafted local features such as SIFT 26 and SURF 27 or global features such as HOG 28 are now frequently replaced by learned features 29 .”
Hyperspectral dataset of historical documents and mock-ups from 400 to 1700 nm (HYPERDOC)
López-Baldomero, A. B., Nieves, J. L., Moronta-Montero, F., et al. (2025) · Scientific Data · doi:10.1038/s41597-025-05599-0 · Cited by 5
Robust Vehicle Detection in Aerial Images Based on Image Spatial Pyramid Detection Model
Li, X., & Li, X. (2019) · 2019 IEEE 4th International Conference on Advanced Robotics and Mechatronics (ICARM) · doi:10.1109/icarm.2019.8834183 · Cited by 5
“Therefore, the SURF (Speeded Up Robust Feature) [18] is utilized to remove the patches which contain few vehicles with great probability.”
MethodEfficient Robot Localization Through Deep Learning-Based Natural Fiduciary Pattern Recognition
Almonte, R. A. M., Zulueta, E., Etxeberria‐Agiriano, I., & Fernández‐Gámiz, U. (2025) · Mathematics · doi:10.3390/math13030467 · Cited by 5
Wei, X., Wang, X., Wang, X., & Du, Y. (2025) · Structures · doi:10.1016/j.istruc.2025.108714 · Cited by 6
Hybrid Deep Learning Framework for Eye-in-Hand Visual Control Systems
Botezatu, A.-P., Iancu, A.-I., & Burlacu, A. (2025) · Robotics · doi:10.3390/robotics14050066 · Cited by 4
“The proposed approach examines the effect of early fusion on the performance of the SURF (Speed-Up Robust Features) [30,31] operator.”
VD-Matcher: A Very Deep Local Feature Matcher With Weight Recycling and Keypoint Detection
Dai, K., Zhou, Z., Jiang, Z., Sun, Q., Xie, T., Gao, H., An, T., Li, R., & Zhao, L. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3574735 · Cited by 9
“The detector-based pipeline, extensively explored in prior researches [5], [6], [7], [8], [9], [10], [11], follows a standard paradigm that includes feature detection, description, and matching.”
Learning a Perspective-Invariant Descriptor for Remote Sensing Image Matching
Wang, J., Qu, Z., Kong, L., Yuan, W., Liu, E., Zhang, R., & Fu, R. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3574761 · Cited by 6
“Some modifications replace the gradient orientation used in SIFT with alternative features, such as SURF [14], CS-LBP [15], and Gabor-SIFT [16].”
Zhang, C., Zhu, J., Ma, Y., & Huang, Q. (2025) · Scientific Reports · doi:10.1038/s41598-025-13435-z · Cited by 6
Impact of Underwater Image Enhancement on Feature Matching
Summers, J. M., Jones, M. W., & Seale, C. (2025) · Sensors · doi:10.3390/s25226966 · Cited by 8
“FUnIE-GAN [43] is a successful example of the utilisation of a multitude of meth-ods.”
SPT: Obtaining long trajectory data of disordered traffic using a swarm of unmanned aerial vehicles
Rajput, S., Venkateshappa, S., Kanagaraj, V., Asaithambi, G., & Treiber, M. (2025) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2025.105431 · Cited by 10
Jing, W., Kang, P., Di, D., Wang, J., Song, Y., Li, C., & Fan, L. (2026) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2025.12.021 · Cited by 5
Melvin, S. A., Wang, H., & Yazdani, A. (2026) · Robotica · doi:10.1017/s0263574725102518 · Cited by 6
“Modern VSLAM methods based on point features typically employ feature detection algorithms like SIFT [35], SURF [77], and Oriented FAST and Rotated BRIEF (ORB) [78] to extract key points for matching images and then determined the pose by minimizing reprojection error.”
Zhang, S., Huang, Y., Zhao, W., Zhao, W., Guan, Z., & Peng, J. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00812 · Cited by 4
“Early approaches rely on local or global features to establish correspondences between keypoints on CAD models [1, 3, 4].”
Monocular target detection on transport infrastructures with dynamic and variable environments
Alvarez, S., Llorca, D. F., Sotelo, M. Á., et al. (2012) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2012.6338690 · Cited by 4
“It captures the movement of static feature points (SURF [18]) between the current image and the background model, to estimate the camera displacement.”
BackgroundLi, L., Liu, Y., & Wang, K. (2013) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio.2013.6739616 · Cited by 4
“The Omnidirectional Projection Model There are different designs in the centra vision system and dioptric vision system system as a sample to derive the projection model and it is easy to extend the results to the dioptric one based on the unifying theory of omnidirectional vision [23], [24].”
BackgroundUnderwater Photogrammetry and Visual Odometry
Royer, J., Pasquet, J., Merad, D., & Drap, P. (2018) · MDPI eBooks · doi:10.3390/books978-3-03842-685-1-12 · Cited by 5
“Scale invariant feature descriptors (SIFT) Lowe (2004) and the Speeded Up Robust Features (SURF) Bay et al. (2006) are well used in this context Nawaf and Tremeau (2014).”
MethodA Hierarchical Utilization of Semantic Gradients and Scene Structure for Visual Place Recognition
Bao, Y., Pan, Y., Yang, Z., & Huan, R. (2023) · IEEE Transactions on Cognitive and Developmental Systems · doi:10.1109/tcds.2023.3281870 · Cited by 3
“Traditional VPR systems are based either on local features, such as handcrafted features like SIFT [3] and SURF [4], or global features, such as HOG [5] and Gist [6].”
MethodLifeSeeker: an interactive concept-based retrieval system for lifelog data
Nguyen, T.-N., Le, T.-K., Ninh, V.-T., et al. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15317-w · Cited by 4
SiLK-SLAM: accurate, robust and versatile visual SLAM with simple learned keypoints
Yao, J., & Li, Y. (2024) · Industrial Robot the international journal of robotics research and application · doi:10.1108/ir-11-2023-0309 · Cited by 4
Robust corner detection in continuous space
Wei, X., Dong, Y., Liu, Q., Wang, L., & Lou, L. (2024) · The Visual Computer · doi:10.1007/s00371-024-03362-x · Cited by 4
“The comparison methods include Harris [5], FAST [17], SURF [19], SIFT b [16], and the method of He-Yung c [34].”
Mao, H., Ma, Y., Zhang, D., et al. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3467256 · Cited by 4
Walz, J., Hunhold, P., & Linß, E. (2024) · Machine learning for computational science and engineering · doi:10.1007/s44379-024-00002-y · Cited by 5
“GLCM [24] and the Haralick [25] computation were chosen from the available texture features.”
The Ground-Penetrating Radar Image Matching Method Based on Central Dense Structure Context Features
Xu, J., Lai, Q., Wei, D., et al. (2024) · Remote Sensing · doi:10.3390/rs16224291 · Cited by 4
SDE2D: Semantic-Guided Discriminability Enhancement Feature Detector and Descriptor
Li, J., Zhang, R., Li, G., & Li, T. H. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3521748 · Cited by 6
Hierarchical loop closure detection with weighted local patch features and global descriptors
Ren, M., Zhang, X., Liu, B., & Zhu, Y. (2025) · Applied Intelligence · doi:10.1007/s10489-024-06135-0 · Cited by 4
LoFLAT: Local Feature Matching using Focused Linear Attention Transformer
Cao, N., He, R., Dai, Y., et al. (2024) · Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) · doi:10.1109/apsipaasc63619.2025.10848757 · Cited by 4
“Meth-ods such as SIFT [6], SURF [7], and ORB [8] exemplify this category, offering high robustness and accuracy in identifying correspondences in structured and well-textured environments.”
BackgroundExtending 6D Object Pose Estimators for Stereo Vision
Pöllabauer, T., Emrich, J., Knauthe, V., & Kuijper, A. (2025) · Lecture notes in computer science · doi:10.1007/978-981-97-8705-0_8 · Cited by 7
Robust Image Registration via Consistent Topology Sort and Vision Inspection
Yang, J. H., Han, P., Huang, J. H., Li, Q., Wang, C., & Li, X. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3556000 · Cited by 5
“In recent decades, numerous feature-based methods are proposed, e.g., the typical scale-invariant feature transform (SIFT) [24], its variant speeded up robust features (SURFs) [25], FAST corner [26], and rotated BRIEF (ORB) [27], which have been successfully applied to image stitching task.”
Liu, B., Qian, Y., & Wang, J. (2025) · Tsinghua Science & Technology · doi:10.26599/tst.2024.9010117 · Cited by 4
A database for automatic identification of herbarium specimens in Piperaceae family
Kajihara, A. Y., de Queiróz, G. A., Caxambú, M. G., Oliveira, L. S., Bertolini, D. S., & Schwerz, A. L. (2025) · Multimedia Tools and Applications · doi:10.1007/s11042-025-20883-2 · Cited by 5
“Only non-handcrafted feature extractors were evaluated in the RGB mode since LBP [56] and SURF [59] are designed for grayscale images.”
LIM: Lightweight Image Local Feature Matching
Ying, S., Zhao, J., Li, G., & Dai, J. (2025) · Journal of Imaging · doi:10.3390/jimaging11050164 · Cited by 6
“To address this, researchers have introduced alternative approaches such as FAST (Features from Accelerated Segment Test) [6] and SURF (Speeded-Up Robust Features) [7], which aim to balance feature extraction effectiveness with reduced computational complexity.”
Cai, W., Huang, L., & Zou, Z. (2025) · Journal of Computing in Civil Engineering · doi:10.1061/jccee5.cpeng-6376 · Cited by 6
A novel multi-camera DIC without point cloud stitching using global co-visible relationships
Zhu, K., Chi, Y., Zhao, J., et al. (2025) · Optics & Laser Technology · doi:10.1016/j.optlastec.2025.113527 · Cited by 4
Vision-based geometric finite element model updating for cable suspension bridges
Lee, Y.-W., Lee, J. H., Byun, N., et al. (2025) · Advanced Engineering Informatics · doi:10.1016/j.aei.2025.103714 · Cited by 4
Geometry-Constrained Monocular Scale Estimation Using Semantic Segmentation for Dynamic Scenes
Zhang, H., Wu, Z., Li, H., Shangguan, Q., & An, K. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3625332 · Cited by 5
“Traditional ones like SIFT [53], [54], SURF [55], BRISK [56] can detect scale-invariant and robust key points, and ORB [24] boosts e ffi ciency and rotational invariance.”
Iterative Versus Voting Method to Reach Consensus Given Multiple Correspondences of Two Sets
Moreno‐García, C. F., Cortés, X., & Serratosa, F. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-19390-8_60 · Cited by 4
A Cognitive Approach for Robots’ Vision Using Unsupervised Learning and Visual Saliency
Ramík, D. M., Sabourin, C., & Madani, K. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-21501-8_11 · Cited by 4
“Speed-up Robust Features, or SURF, described in [5] is a well-established image detector technique based on matching interest points on the source image with interest points coming from the template.”
BackgroundMethodCOGE: A Novel Binary Feature Descriptor Exploring Anisotropy and Non-uniformity
Mao, Z., Zhang, Y., & Tian, Q. (2013) · Lecture notes in computer science · doi:10.1007/978-3-319-03731-8_34 · Cited by 4
“proposed the SURF descriptor [2].”
Multi-camera Vehicle Tracking Using Local Image Features and Neural Networks
Dalka, P. (2012) · Communications in computer and information science · doi:10.1007/978-3-642-30721-8_6 · Cited by 4
A Biologically Inspired Approach for Fast Image Processing
Coleman, S., Scotney, B. W., & Gardiner, B. (2013) · International Conference on Machine Vision · Cited by 4
A comparison of two algorithms for robot learning from demonstration
Suay, H. B., & Chernova, S. (2011) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/icsmc.2011.6084052 · Cited by 4
The Design of a Hybrid Feature Detector for Adult Images
Tsai, M.-J., & Chang, H.-S. (2013) · IEEE International Conference on Semantic Computing (ICSC) · doi:10.1109/icsc.2013.73 · Cited by 4
“Therefore, some other researches [7-8] have proposed BoVWs based technique, which is popular for object recognition and visual categorization [9-12].”
BackgroundMethodCGCI-SIFT: A More Efficient and Compact Representation of Local Descriptor
Su, D., Wu, J., Cui, Z., Sheng, V. S., & Gong, S. (2013) · Measurement Science Review · doi:10.2478/msr-2013-0022 · Cited by 5
“The thresholdβ is varied to obtain the curves to present our experimental results (recall versus 1-precision, which is used in [6], [7] and [17]) in Section 4.”
MethodAutonomous detection of information patterns through hierarchical peeling
Radványi, M., & Karacs, K. I. (2013) · European Conference on Circuit Theory and Design · doi:10.1109/ecctd.2013.6662253 · Cited by 4
Local region statistical distance measure for tracking in Wide Area Motion Imagery
Mathew, A., & Asari, V. K. (2012) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/icsmc.2012.6377708 · Cited by 4
Orientation-aware scene understanding for mobile cameras
Wang, J., Schindler, G., & Essa, I. A. (2012) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/2370216.2370258 · Cited by 4
3DComplete: Efficient completeness inspection using a 2.5D color scanner
So, E. W. Y., Munaro, M., Michieletto, S., et al. (2013) · Computers in Industry · doi:10.1016/j.compind.2013.03.014 · Cited by 4
PASU: A personal area situation understanding system using wireless camera sensor networks
Yoon, S., Oh, H., Lee, D., et al. (2012) · Personal and Ubiquitous Computing · doi:10.1007/s00779-012-0611-5 · Cited by 4
“As done in [3], we use a support vector machine (SVM) classifier to determine whether a HOG descriptor represents a person or not, and Joachims’ implementation of SVMlight [14] is ported to the CITRIC board.”
A MapReduce Approach for SIFT Feature Extraction
Han, W., Kang, Y., Yang, C., & Zhang, X. (2013) · doi:10.1109/cloudcom-asia.2013.22 · Cited by 4
Combining feature descriptors for efficient classification and recognition of images
Roy, K., Subrahmanya, G., & Rao, V. S. (2012) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp.2012.6470022 · Cited by 4
“Popular keypoint based features include Scale-invariant feature transform (SIFT) by Lowe [18], Speeded-Up Robust Features (SURF) suggested by Herbert Bay [19].”
MethodDetermination of platform attitude through SURF based aerial image matching
Jing, L., Xu, L., Li, X., et al. (2013) · IEEE International Conference on Imaging Systems and Techniques (IST) · doi:10.1109/ist.2013.6729654 · Cited by 4
Metadata
- Authors
- Bay, H., Tuytelaars, T., & Van Gool, L.
- Affiliations
- ETH Zurich (CH)
- KU Leuven (BE)
- ORCID
- Tuytelaars, Tinne
- Venue
- Lecture Notes in Computer Science
- Publisher
- Springer Berlin Heidelberg
- Published
- 2006
- Pages
- 404-417
- ISSN
- 0302-9743 (ISSN-L)
- 1611-3349
- OpenAlex
- W1677409904
- Semantic Scholar
- 9ad3179098eb9ac7f0c81bf89b6bdcde91752880
- Cited by
- 14,955
- References
- 39
- Citations by year
- 2026: 210 · 2025: 519 · 2024: 727 · 2023: 757 · 2022: 878 · 2021: 975 · 2020: 978 · 2019: 1036 · 2018: 968 · 2017: 976 · 2016: 930 · 2015: 1079 · 2014: 1006 · 2013: 982 · 2012: 884
- FWCI
- 196.80
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- www.springer.com/tdm (tdm)
- Locations
- Lecture notes in computer science (publishedVersion)
- Lirias (KU Leuven) (submittedVersion, other-oa)
- Funders
- Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image Retrieval and Classification Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Object Detection Techniques
- Keywords
- SURF, keypoint detection, feature descriptors, scale invariance, rotation invariance, integral image, Hessian matrix, feature matching, object recognition, repeatability
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.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
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.