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 2,601–2,700 of 14,752
Vetrella, A. R., Opromolla, R., Fasano, G., et al. (2017) · AIAA Information Systems-AIAA Infotech @ Aerospace · doi:10.2514/6.2017-0879 · Cited by 27
Texture-less object detection and 6D pose estimation in RGB-D images
Zhang, H., & Cao, Q. (2017) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2017.06.003 · Cited by 29
A comprehensive evaluation of local detectors and descriptors
Wu, S., Oerlemans, A., Bakker, E. M., & Lew, M. S. (2017) · Signal Processing Image Communication · doi:10.1016/j.image.2017.06.010 · Cited by 32
Augmented songbook: an augmented reality educational application for raising music awareness
Rusiñol, M., Chazalon, J., & Díaz-Chito, K. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-4991-4 · Cited by 40
GPU Accelerated FFT-Based Registration of Hyperspectral Scenes
Ordóñez, Á., Argüello, F., & Heras, D. B. (2017) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2017.2734052 · Cited by 30
Multi-Frame Super-Resolution of Gaofen-4 Remote Sensing Images
Xu, J., Liang, Y., Liu, J., et al. (2017) · Sensors · doi:10.3390/s17092142 · Cited by 27
Improving CBIR accuracy using convolutional neural network for feature extraction
Shah, A., Naseem, R., Sadia, Iqbal, S., & Shah, M. A. (2017) · International Conference on Emerging Technologies · doi:10.1109/icet.2017.8281730 · Cited by 33
Guo, Q., Xiao, J., & Hu, X. (2018) · Sensors · doi:10.3390/s18030698 · Cited by 27
ElTantawy, A., & Shehata, M. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2843761 · Cited by 29
Davis, J., & Pernicka, H. J. (2018) · Acta Astronautica · doi:10.1016/j.actaastro.2018.10.033 · Cited by 28
Bawa, V. S., & Kumar, V. (2018) · Neural Computing and Applications · doi:10.1007/s00521-018-3867-5 · Cited by 30
“For comparative analysis, we considered scene feature extractor like scale-invariant feature transform (SIFT) [27], Speeded Up Robust Features (SURF) [2] and Oriented FAST, Oriented BRIEF (ORB) [20] and CENTRIST [29].”
MethodInfluentialDetection of abnormalities in mammograms using deep features
Tavakoli, N., Karimi, M., Norouzi, A., Karimi, N., Samavi, S., & Soroushmehr, S. M. R. (2019) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-019-01639-x · Cited by 27
“Many methods use CNN to detect abnormal areas in mammographic images (Jiao et al. 2015; Ertosun and Rubin 2015; Jaffar 2017; Abbas 2016; Bay et al. 2006; Guo et al. 2010; Schmidhuber 2015; Qiu et al. 2016; Lo et al. 2002; Zhang et al. 2018).”
BackgroundMethodDepth Sensing Beyond LiDAR Range
Zhang, K., Xie, J., Snavely, N., & Chen, Q. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.00176 · Cited by 29
“We first detect sparse matches between the left and back images with SURF [4].”
Influential3D Lunar Terrain Reconstruction from Apollo Images
Broxton, M. J., Nefian, A., Moratto, Z. M., et al. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-10331-5_66 · Cited by 27
Bhattacharya, S., Mehran, R., Sukthankar, R., & Shah, M. A. (2014) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2014.2300833 · Cited by 28
Facet-based multiple building analysis for robot intelligence
Trinh, H.-H., Kim, D.-N., & Jo, K.-H. (2008) · Applied Mathematics and Computation · doi:10.1016/j.amc.2008.05.059 · Cited by 37
Vision-Based System for Human Detection and Tracking in Indoor Environment
Benezeth, Y., Emile, B., Laurent, H., & Rosenberger, C. (2009) · International Journal of Social Robotics · doi:10.1007/s12369-009-0040-4 · Cited by 31
FXPAL Interactive Search Experiments for TRECVID 2007
Pickens, J., Adcock, J. E., Cooper, M., & Girgensohn, A. (2008) · TRECVID · Cited by 26
Ngeljaratan, L. N., Baş, E. E., & Moustafa, M. A. (2024) · Sensors · doi:10.3390/s24051450 · Cited by 31
DexiTac: Soft Dexterous Tactile Gripping
Lu, C., Tang, K., Yang, M., Yue, T., Li, H., & Lepora, N. F. (2024) · IEEE/ASME Transactions on Mechatronics · doi:10.1109/tmech.2024.3384432 · Cited by 27
“The positions ( x m , y m ) of the M tactile markers on the image are then detected using Determinant of Hessian (DoH) blob recognition [25] (see Fig.5 ).”
MethodA comparative analysis of pairwise image stitching techniques for microscopy images
Mohammadi, F. S., Mohammadi, S. E., Adi, P. M., Mirkarimi, S. M. A., & Shabani, H. (2024) · Scientific Reports · doi:10.1038/s41598-024-59626-y · Cited by 29
“To implement pairwise registration using phase correlation and NCC we utilized the pairwise registration part of the Microscopy Image Stitching Tool (MIST) source code 24 .”
MethodYang, X., Bai, B., Zhang, Y., Aydın, M., Li, Y., Selçuk, Ş. Y., Costa, P. C., Guo, Z., Fishbein, G. A., Atlan, K., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-52263-z · Cited by 36
Neural Clustering Based Visual Representation Learning
Chen, G., Li, X., Yang, Y., & Wang, W. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.00546 · Cited by 43
Kılıç, Ş. (2025) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-025-01479-6 · Cited by 29
“Feature extraction algorithms like Speeded Up Robust Features (SURF) [7], Scale Invariant Feature Transform (SIFT) [8], Histogram of Oriented Gradients (HOG) [9], and Grey Level Co-occurrence Matrices (GLCM) [10] have been used with moderate success, but there are still some limitations.”
Multi-Stage Video Analysis Framework
Czyewski, A., Szwoch, G., Dalka, P., et al. (2011) · InTech eBooks · doi:10.5772/16088 · Cited by 26
“Final vehicle feature vector based on SURF descriptors contains 128 × 4 = 512 elements.”
BackgroundInfluentialAffine Subspace Representation for Feature Description
Wang, Z., Fan, B., & Wu, F. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-10584-0_7 · Cited by 30
Loosely Distinctive Features for Robust Surface Alignment
Albarelli, A., Rodolà, E., & Torsello, A. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-15555-0_38 · Cited by 29
“Among the most used descriptors are the Scale-invariant feature transform (SIFT) [5], the Speeded Up Robust Features (SURF) [6] and Gradient Location and Orientation Histogram (GLOH) [7].”
MethodSURF algorithm implementation on FPGA
Sledević, T., & Serackis, A. (2012) · Proceedings of the ... biennial Baltic Electronics Conference · doi:10.1109/bec.2012.6376874 · Cited by 26
“Wavelets are calculated for pixels that are located within a radius of 6 around the interest point [1].”
BackgroundMethodInfluentialVision-based loop closing for delayed state robot mapping
Ila, V., Andrade‐Cetto, J., Valencia, R., & Sanfeliu, A. (2007) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2007.4399221 · Cited by 33
Video-based localization without 3D mapping for the visually impaired
Liu, J. J., Phillips, C. J., & Daniilidis, K. (2010) · 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2010.5543581 · Cited by 26
“that uses Hessian matrix-based measure for detection and a distribution-based descriptor [25].”
MethodThick boundaries in binary space and their influence on nearest-neighbor search
Trzcinski, T. P., Lepetit, V., & Fua, P. (2012) · Pattern Recognition Letters · doi:10.1016/j.patrec.2012.08.006 · Cited by 33
Multiview Face Detection and Registration Requiring Minimal Manual Intervention
Anvar, S. M. H., Yau, W., & Teoh, E. K. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2013.37 · Cited by 26
., P. A. (2015) · International Journal of Research in Engineering and Technology · doi:10.15623/ijret.2015.0402052 · Cited by 28
Li, Y., Dong, W., Shi, G., et al. (2015) · IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.59 · Cited by 26
“Specifically, the SURF [2, 1] feature descriptors are used for image retrieval, and the method of [13] is employed to retrieve the correlated images.”
MethodCloud Computing and Augmented Reality for Cultural Heritage
Vecchio, P., Mele, F., Paolis, L. T. D., et al. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-22888-4_5 · Cited by 26
Evaluation of Feature Detection in HDR Based Imaging Under Changes in Illumination Conditions
Rana, A., Valenzise, G., & Dufaux, F. (2015) · IEEE International Symposium on Multimedia (ISM) · doi:10.1109/ism.2015.58 · Cited by 26
“For blob detection, we carried out experiments with the highly robust SURF [4] detector.”
MethodInfluentialZhi, X., Yan, J., Hang, Y., & Wang, S. (2016) · Journal of Real-Time Image Processing · doi:10.1007/s11554-016-0594-y · Cited by 26
“Constructing and matching descriptors in SIFT-based [1] (SIFT means ‘‘Scale-Invariant Feature Transform’’, and it’s an algorithm to detect and describe local features) or SURF-based [2] (SURF means ‘‘Speeded Up Robust Features’’, and it’s a local feature detector and descriptor partly inspired by SIFT) methods are ver…”
Object-Centric Representation Learning from Unlabeled Videos
Gao, R., Jayaraman, D., & Grauman, K. L. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-54193-8_16 · Cited by 31
Automatic measurement of anthropometric dimensions using frontal and lateral silhouettes
Aslam, M., Rajbdad, F., Khattak, S., & Azmat, S. (2017) · IET Computer Vision · doi:10.1049/iet-cvi.2016.0406 · Cited by 27
“Similarly, some other techniques such as Scale Invariant Feature Transform (SIFT) [26], Harris-Laplacian [27], Speeded-Up Robust Features (SURF) [28] can be used for human body FPs detection.”
MethodPlace Recognition: An Overview of Vision Perspective
Zeng, Z., Zhang, J., Wang, X., Chen, Y., & Zhu, C. (2018) · Applied Sciences · doi:10.3390/app8112257 · Cited by 30
“At early stages, place recognition was dominated by sophisticated local-invariant feature extractors such as SIFT Lowe (2004) and SURF Bay et al. (2006), hand-crafted global image descriptors such as GIST Oliva and Torralba (2001; 2006), and bag-of-visual-words Philbin et al. (2007); Sivic et al. (2003) approach.”
MethodInfluentialOn Combining Social Media and Spatial Technology for POI Cognition and Image Localization
Qian, X., Lu, X.-Q., Han, J.-W., Du, B., & Li, X.-L. (2017) · Proceedings of the IEEE · doi:10.1109/jproc.2017.2731600 · Cited by 22
“The SIFT feature is often utilized [36], which also has some simplified forms such as SURF [39].”
MethodCopy move forgery detection based on keypoint and patch match
Liu, K., Lu, W., Lin, C., Huang, X., Liu, X., Yeung, Y., & Xue, Y. (2019) · Multimedia Tools and Applications · doi:10.1007/s11042-019-07930-5 · Cited by 31
Maiwald, F., Bruschke, J., Lehmann, C. U., & Niebling, F. (2019) · Virtual Archaeology Review · doi:10.4995/var.2019.11867 · Cited by 30
Entropy-Based Face Recognition and Spoof Detection for Security Applications
Pujol, F. A., Pujol, M. J., Rizo-Maestre, C., & López, M. P. (2019) · Sustainability · doi:10.3390/su12010085 · Cited by 25
“Among all local feature descriptors, the most commonly used are SIFT (Scale-Invariant Feature Transform) [24,25], HOG (Histogram of Oriented Gradients) [26,27], SURF (Speeded-up Robust Features) [28,29] and LBP (Local Binary Patterns) [30,31], which are used to address variability in the image caused by changes in per…”
Advance on large scale near-duplicate video retrieval
Shen, L., Hong, R., & Hao, Y. (2020) · Frontiers of Computer Science · doi:10.1007/s11704-019-8229-7 · Cited by 25
“The commonly used lo-cal features include scale invariant feature transform (SIFT) [56–59], SURF [13, 60, 61], and the local binary pattern (LBP) [52,62–64], etc. 1) SIFT.”
MethodExplaining away results in accurate and tolerant template matching
Spratling, M. (2020) · Pattern Recognition · doi:10.1016/j.patcog.2020.107337 · Cited by 28
“Many alternative feature descriptors have also been proposed, such as SURF (Bay et al., 2006), BRIEF (Calonder et al.”
Smart Shopping Carts Based on Mobile Computing and Deep Learning Cloud Services
Sarwar, M. A., Daraghmi, Y., Liu, K.-W., Chi, H.-C., İk, T.-U., & Li, Y.-L. (2020) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc45663.2020.9120574 · Cited by 38
“…detection [13], the scale-invariant feature transform (SIFT) algorithm was utilized for the detection of the local feature of images [14], the speeded-up robust features (SURF) is the descriptor and local features detector utilized for classification, object detection, and image recognition [15].”
MethodAn approach for underwater image enhancement based on color correction and dehazing
Zhang, Y., Yang, F., & He, W. (2020) · International Journal of Advanced Robotic Systems · doi:10.1177/1729881420961643 · Cited by 30
Semantic Structure from Motion for Railroad Bridges Using Deep Learning
Park, G., Lee, J. H., & Yoon, H. (2021) · Applied Sciences · doi:10.3390/app11104332 · Cited by 26
Detecting Stable Keypoints from Events through Image Gradient Prediction
Chiberre, P., Perot, E., Sironi, A., & Lepetit, V. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00153 · Cited by 32
Image classification with multi-view multi-instance metric learning
Tang, J., Li, D., & Tian, Y. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.116117 · Cited by 27
RetinaNet: A deep learning architecture to achieve a robust wake detector in SAR images
Prete, R. D., Graziano, M. D., & Renga, A. (2021) · IEEE International Forum on Research and Technology for Society and Industry (RTSI) · doi:10.1109/rtsi50628.2021.9597297 · Cited by 26
Semi-Supervised 6D Object Pose Estimation Without Using Real Annotations
Zhou, G., Wang, D., Yan, Y., Chen, H., & Chen, Q. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3138129 · Cited by 31
FAST, BRIEF and SIFT based image copy-move forgery detection technique
Fatima, B., Ghafoor, A., Ali, S. S., & Riaz, M. M. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-12915-y · Cited by 26
Optimal feature selection and classification of Indian classical dance hand gesture dataset
Raj, R., Dharan, S., & Sunil, T. T. (2022) · The Visual Computer · doi:10.1007/s00371-022-02572-5 · Cited by 29
BDA-SketRet: Bi-level domain adaptation for zero-shot SBIR
Chaudhuri, U., Chavan, R., Banerjee, B., Dutta, A., & Akata, Z. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.09.104 · Cited by 28
Image-based geometric digital twinning for stone masonry elements
Pantoja-Rosero, B. G., Saloustros, S., Achanta, R., & Beyer, K. (2022) · Automation in Construction · doi:10.1016/j.autcon.2022.104632 · Cited by 27
Learning to Reduce Scale Differences for Large-Scale Invariant Image Matching
Fu, Y., Zhang, P., Liu, B., Zheng, R. Y., & Wu, Y. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3210602 · Cited by 35
“The Multi-Scale Feature Extraction method only extracts local features from the scale levels near original image scale levels [5], [10], [12], [20].”
MethodSemi-Direct Multimap SLAM System for Real-Time Sparse 3-D Map Reconstruction
Xie, H., Zhang, D., Wang, J., Zhou, M., Cao, Z., Hu, X., & Abusorrah, A. M. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3240206 · Cited by 27
“Compared with scale invariant feature transform (SIFT) [50] and speeded up robust features (SURF) [51], the ORB method can achieve a huge improvement in computational efficiency while ensuring rotation and scale invariance.”
MethodStorytelling with Image Data: A Systematic Review and Comparative Analysis of Methods and Tools
Lotfi, F., Beheshti, A., Farhood, H., Pooshideh, M., Jamzad, M., & Beigy, H. (2023) · Algorithms · doi:10.3390/a16030135 · Cited by 24
“To be able to detect the objects without the rotation and scaling concerns, intersect point and local features detectors such as Scale-Invariant Feature Transform (SIFT) [42] and Speeded Up Robust Features (SURF) [43] could be used.”
Efficiency Matters: Speeding Up Automated Testing with GUI Rendering Inference
Feng, S., Xie, M., & Chen, C. (2023) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1109/icse48619.2023.00084 · Cited by 39
“In detail, we adopted three types of feature extraction methods used in machine learning, e.g., Scale invariant feature transform (SIFT) [46], Speed up robot features (SURF) [47], and Oriented fast and rotated brief (ORB) [48].”
BackgroundEvaluation of feature detection algorithms for structure from motion
Govender, N. (2009) · Cited by 35
Gaussian Process Gauss-Newton for 3D laser-based Visual Odometry
Tong, C. H., & Barfoot, T. D. (2013) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2013.6631321 · Cited by 29
Rapid scene reconstruction on mobile phones from panoramic images
Pan, Q., Arth, C., Reitmayr, G., Rosten, E., & Drummond, T. W. (2011) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar.2011.6092370 · Cited by 27
Object Recognition on a Chip: A Complete SURF-Based System on a Single FPGA
Schaeferling, M., & Kiefer, G. (2011) · International Conference on ReConFigurable Computing and FPGAs (ReConFig) · doi:10.1109/reconfig.2011.65 · Cited by 26
“Presently available algorithms such as SIFT or SURF can to some extent solve these problems by computing so-called point features, which are invariant towards scaling and rotation [1][4][11][16].”
BackgroundMethodComputing Optimised Parallel Speeded-Up Robust Features (P-SURF) on Multi-Core Processors
Zhang, N. (2009) · International Journal of Parallel Programming · doi:10.1007/s10766-009-0122-9 · Cited by 29
“It was first introduced in [11] and later more fully explained in [12].”
BackgroundA tracking framework for augmented reality tours on cultural heritage sites
Seo, B.-K., Kim, K., Park, J., & Park, J.-I. (2010) · ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry (VRCAI) · doi:10.1145/1900179.1900215 · Cited by 31
“For computational ef.ciency, however, they are tracked using optical .ow analysis [Bouguet 2000] and SURF is performed for initialization or recovery of a pose.”
MethodInfluentialSegmentation-based multi-class semantic object detection
Vieux, R., Benois‐Pineau, J., Domenger, J., & Braquelaire, A. J. P. (2010) · Multimedia Tools and Applications · doi:10.1007/s11042-010-0611-2 · Cited by 25
Visual precis generation using coresets
Paul, R., Feldman, D., Rus, D. L., & Newman, P. M. (2014) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2014.6907021 · Cited by 24
“A visual vocabulary [12] of approximately 11k visual words was generated by clustering SURF features [21] extracted from this data set.”
MethodA Multiple Kernel Learning Approach to Joint Multi-class Object Detection
Lampert, C. H., & Blaschko, M. B. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-69321-5_4 · Cited by 28
Combined intention, activity, and motion recognition for a humanoid household robot
Gehrig, D., Krauthausen, P., Rybok, L., Kuehne, H., Hanebeck, U. D., Schultz, T., & Stiefelhagen, R. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6095118 · Cited by 38
“The location and spatial size of the gradient ROI is determined by employing a fast 2D interest point detector [25] to every fourth frame.”
MethodLogo detection based on spatial-spectral saliency and partial spatial context
Gao, K., Lin, S., Zhang, Y., Tang, S., & Zhang, D. (2009) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2009.5202500 · Cited by 30
“For a given query logo, the same processing will be used to obtain its SURF and spatial information, and all of these features are stored in an effective index something like k-d tree.”
MethodInfluentialMAV visual SLAM with plane constraint
Lee, G. H., Fraundorfer, F., & Pollefeys, M. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5980365 · Cited by 28
A comparison of feature detectors and descriptors for object class matching
Hietanen, A., Lankinen, J., Kämäräinen, J., Buch, A. G., & Krüger, N. (2015) · Neurocomputing · doi:10.1016/j.neucom.2015.08.106 · Cited by 31
A Tool to Support the Creation of Datasets of Tampered Videos
Ardizzone, E., & Mazzola, G. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-23234-8_61 · Cited by 26
“Of course, if a homogenous area is selected, our method fails, as no SURF points can be extracted and no tracking can be performed, not even with other algorithms.”
BackgroundInfluentialDantone, M., Bossard, L., Quack, T., & Gool, L. V. (2011) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2011.6130218 · Cited by 26
Self-citationFinding local leaf vein patterns for legume characterization and classification
Larese, M. G., & Granitto, P. M. (2015) · Machine Vision and Applications · doi:10.1007/s00138-015-0732-8 · Cited by 30
“One direction may be to compare the vein patches obtained by SIFT with other state-of-the-art feature detectors and descriptors, such as Speeded-Up Robust Features (SURF) [3].”
BackgroundComparative study of descriptors with dense key points
Chatoux, H., Lecellier, F., & Fernández-Maloigne, C. (2016) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2016.7899928 · Cited by 26
“SURF and FREAK measure the angle in the descriptor chain and not in the detector one, so pre-giving an angle has no effect on their results.”
MethodInfluentialA Simplified Method for UAV Multispectral Images Mosaicking
Ren, X., Sun, M., Zhang, X., et al. (2017) · Remote Sensing · doi:10.3390/rs9090962 · Cited by 26
Deep learning features exception for cross-season visual place recognition
Kenshimov, C., Bampis, L., Amirgaliyev, B., Arslanov, M., & Gasteratos, A. C. (2017) · Pattern Recognition Letters · doi:10.1016/j.patrec.2017.10.028 · Cited by 24
Interactive Mobile Augmented Reality System for Image and Hand Motion Tracking
Chiu, P.-H., Tseng, P., & Feng, K. (2018) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2018.2864893 · Cited by 29
“Different feature detection methods have been proposed such as features from accelerated segment test (FAST) corner detection [7], scale-invariant feature detection (SIFT) [8], and speeded-up robust features (SURF) detection scheme [9].”
MethodSSIR: Secure similarity image retrieval in IoT
Yan, H., Chen, Z., & Jia, C. (2018) · Information Sciences · doi:10.1016/j.ins.2018.11.046 · Cited by 25
Exploiting Multi-layer Features Using a CNN-RNN Approach for RGB-D Object Recognition
Çağlayan, A., & Can, A. B. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-11015-4_51 · Cited by 31
CDbin: Compact Discriminative Binary Descriptor Learned With Efficient Neural Network
Ye, J., Zhang, S. S., Huang, T., & Rui, Y. (2019) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2896095 · Cited by 32
“SIFT [2] and SURF [3] are widely used hand-crafted descriptors.”
BackgroundMethodCC-NET: Image Complexity Guided Network Compression for Biomedical Image Segmentation
Mishra, S., Liang, P., Czajka, A., et al. (2019) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi.2019.8759448 · Cited by 26
High-Precision Localization Using Ground Texture
Zhang, L., Finkelstein, A., & Rusinkiewicz, S. M. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794106 · Cited by 34
“While SIFT [30] has been popular since its introduction more than a decade ago, more recent alternatives such as SURF [35], ORB [36] have been shown to achieve similar performance at lower computational cost.”
MethodMahmoudi, S. A., Mahmoudi, S. A., Belarbi, M. A., et al. (2020) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.5699 · Cited by 26
Wind turbine tower detection using feature descriptors and deep learning
Abedini, F., Bahaghighat, M., & Shoyan, M. (2020) · Facta universitatis - series Electronics and Energetics · doi:10.2298/fuee2001133a · Cited by 31
ArUco marker-based displacement measurement technique: uncertainty analysis
Tocci, T., Capponi, L., & Rossi, G. L. (2021) · Engineering Research Express · doi:10.1088/2631-8695/ac1fc7 · Cited by 26
Automated Vision-Based Diagnosis of Cassava Mosaic Disease.
Aduwo, J. R., Mwebaze, E., & Quinn, J. A. (2010) · Cited by 26
Maximum Detector Response Markers for SIFT and SURF
Schweiger, F., Zeisl, B., Georgel, P. F., et al. (2009) · Cited by 26
Snap and share your photobooks
Henze, N., & Boll, S. (2008) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/1459359.1459414 · Cited by 25
Rock detection via superpixel graph cuts
Gong, X., & Liu, J. (2012) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2012.6467318 · Cited by 28
“Inspired by the SURF [12] descriptor, we compute the intensity derivates along x and y direction, Ix, Iy , |Ix|, and |Iy| respectively, and choose the mean of each term as a feature to capture texture property.”
MethodKukar, M., Kononenko, I., & Grošelj, C. (2011) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2011.04.009 · Cited by 25
LARGE SCALE URBAN RECONSTRUCTION FROM REMOTE SENSING IMAGERY
Kuschk, G. (2013) · 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-xl-5-w1-139-2013 · Cited by 26
Affine Adaptation of Local Image Features Using the Hessian Matrix
Lakemond, R., Fookes, C., & Sridharan, S. (2009) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2009.8 · Cited by 28
Unmanned Solar Airplanes: Design and Algorithms for Efficient and Robust Autonomous Operation
Leutenegger, S. (2014) · Repository for Publications and Research Data (ETH Zurich) · doi:10.3929/ethz-a-010255301 · Cited by 32
Indexing of the CNN features for the large scale image search
Liu, R., Wei, S., Zhao, Y., & Yang, Y. (2018) · Multimedia Tools and Applications · doi:10.1007/s11042-018-6210-3 · Cited by 28
Deep Neural Networks for Marine Debris Detection in Sonar ImagesPreprint
Valdenegro-Toro, M. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.05241 · Cited by 46
Computer Stereo Vision for Autonomous DrivingPreprint
Fan, R., Wang, L., Bocus, M. J., & Pitas, I. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.03194 · Cited by 27
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.