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 8,901–9,000 of 14,752
Deep Learning for Semiconductor Defect Classification
Sweeney, T., Coleman, S., & Kerr, D. (2022) · IEEE International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin51773.2022.9976162 · Cited by 3
Kujawski, A. (2014) · Logistyka · Cited by 3
Wang, X. (2011) · Springer eBooks · doi:10.1007/978-1-4419-9482-0_3 · Cited by 3
Biologically Inspired Path Execution Using SURF Flow in Robot Navigation
Perez-Sala, X., Ángulo, C., & Escalera, S. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-21498-1_73 · Cited by 3
“D displacements of SURF patterns in the image, where SURF is referred to Speeded Up Robust Features [17].”
MethodRobust Object Recognition in Unstructured Environments
Martínez-Martín, E., & del Pobil, A. P. (2012) · Advances in intelligent systems and computing · doi:10.1007/978-3-642-33926-4_67 · Cited by 3
Machine Learning for Visual Concept Recognition and Ranking for Images
Binder, A., Samek, W., Müller, K., & Kawanabe, M. (2014) · Cognitive technologies · doi:10.1007/978-3-319-06755-1_17 · Cited by 4
Target Localization Based on SIFT Algorithm
Lei, B., Yuanping, J., & Ding, J. (2011) · Communications in computer and information science · doi:10.1007/978-3-642-23214-5_1 · Cited by 3
Viewpoint Invariant Face Detection
Taffar, M., Miguet, S., & Benmohammed, M. (2012) · Communications in computer and information science · doi:10.1007/978-3-642-30567-2_33 · Cited by 3
Augmented reality using spatially multiplexed structured light
Torres, M., Jassel, R., & Tang, Y. (2012) · Cited by 3
Vision-Based Markerless Gaming Interface
Azzari, P., & Stefano, L. D. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-04146-4_32 · Cited by 3
“In particular, the SURF (Speeded Up Robust Features) algorithm [1] reconciled accuracy and low computational cost for robust keypoints extraction and tracking.”
MethodGlobal Matching to Enhance the Strength of Local Intensity Order Pattern Feature Descriptor
Dawood, H., Dawood, H., & Guo, P. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-39065-4_60 · Cited by 3
“[12], in which they got the speed gain due to the usage of integral images, which drastically reduce the number of operations for simple box convolutions, independent of the chosen scale.”
BackgroundCombining Observation Models In Dual Exposure Problems Using The Kullback-Leibler Divergence
Mateos, J., & Tallón, M. (2010) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.42245 · Cited by 3
“The images were photometrically calibrated using histogram equalization and then geometrical calibration was carried out using SURF [5]+RANSAC to obtain the noisy and blurred images shown in Fig.”
MethodVideo stabilization and tampering detection for surveillance systems using homography
Javadi, S., Kadim, Z., Woon, H. H., et al. (2015) · doi:10.1109/i4ct.2015.7219580 · Cited by 3
“[22], is a computationally improved version of the Scale Invariant Feature Transform (SIFT).”
MethodOrdinal pyramid pooling for rotation invariant object recognition
Wang, G., Fan, B., & Pan, C. (2015) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2015.7178190 · Cited by 3
“The first one estimates a dominant orientation of the image patch and calculates descriptor relative to this dominant orientation, e.g. SIFT (Scale Invariant Feature Transform) [6, 7] and SURF (Speeded Up Robust Features) [8].”
MethodPredicting the long-term robustness of visual features
Metka, B., Besetzny, A., Bauer-Wersing, U., & Franzius, M. (2015) · International Conference on Advanced Robotics (ICAR) · doi:10.1109/icar.2015.7251497 · Cited by 3
Visual SLAM Based on Single Omnidirectional Views
Valiente, D., Gil, A., Fernández, L., et al. (2014) · Lecture notes in electrical engineering · doi:10.1007/978-3-319-03500-0_9 · Cited by 3
Robust Spatio-Temporal Features for Human Action Recognition
Mattivi, R., & Shao, L. (2011) · Studies in computational intelligence · doi:10.1007/978-3-642-19551-8_12 · Cited by 3
Automatic Detection of Object of Interest and Tracking in Active Video
Huang, J., & Li, Z.-N. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-10467-1_32 · Cited by 3
Exposing region duplication through local geometrical color invariant features
Gong, J., & Guo, J. (2015) · Journal of Electronic Imaging · doi:10.1117/1.jei.24.3.033010 · Cited by 3
Region Graph Spectra as Geometric Global Image Features
Ho, Q., Yu, W., & Lee, H. K. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-10331-5_24 · Cited by 3
A computed stereoscopic method for laparoscopic surgery by using feature tracking
Lee, P. Y., Yan, S., Hu, M.-H., et al. (2015) · IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan) · doi:10.1109/icce-tw.2015.7217046 · Cited by 3
“We are bit modify the tracking method with SURF technology [7] and applied a depth weights optimization to smooth the depth image.”
MethodSmart device based power generation facility management system in smart grid
Lee, Y.-J., & Kim, E.-K. (2015) · International Conference on Advanced Communication Technology · doi:10.1109/icact.2015.7224919 · Cited by 3
“In this paper, feature point was extracted by using SURF (Speeded Up Robust Features) algorithm so that fast performance ability and feature points that are not changed to rotational change and scale conversion could be found in feature points of panoramic image being transmitted from server and images being obtained…”
MethodAttitude Heading Reference System with Rotation-Aiding Visual Landmarks
Beall, C., Ta, D.-N., Ok, K., et al. (2012) · Cited by 3
Part-Based Method on Handwritten Texts
Wang, S., Uchida, S., & Liwicki, M. (2015) · Cited by 3
Image-based Tracking of Deformable Surfaces
Hilsmann, A., David, C., & Eisert, P. (2011) · InTech eBooks · doi:10.5772/14344 · Cited by 3
Bogen, K. T. (1992) · Fundamental and Applied Toxicology · doi:10.1016/0272-0590(92)90192-k · Cited by 59
Fully affine invariant matching algorithm based on nonlinear scale space for side scan sonar image
Ye, X., Li, P., & Zhang, J. (2015) · IEEE International Conference on Mechatronics and Automation (ICMA) · doi:10.1109/icma.2015.7237860 · Cited by 3
“However, sonar image matching is often very difficult for side scan sonar images because the images acquired from different heading directions usually has large scale and rotation change.”
BackgroundPLEIA: A Reconfigurable Platform for Evaluation of HCI Acting
Peralta, H., Monacelli, É., Riman, C. F., Baklouti, M., Ouezdou, B., Mougharbel, I., Laffont, I., & Bouteille, J. (2007) · doi:10.5772/5173 · Cited by 3
Smart Wheelchair Navigation Based on User’s Gaze on Destination
Tomari, R., Rong, Z., Kobayashi, Y., et al. (2010) · Communications in computer and information science · doi:10.1007/978-3-642-14831-6_52 · Cited by 3
Recognition of stonefish from underwater video
Ashour, H., & Sasi, S. (2015) · International Conference on Advances in Computing, Communications and Informatics (ICACCI) · doi:10.1109/icacci.2015.7275746 · Cited by 3
Efficient parallel Recursive Gaussian SIFT algorithm based on multi-core DSP
Luo, Y., Ye, Z., & Chen, Y. (2015) · IEEE International Conference on Electronics Information and Emergency Communication · doi:10.1109/iceiec.2015.7284568 · Cited by 3
“At execution time, the parallel RGF-SIFT algorithm has higher speedup ratio.”
MethodNotice of Removal Mobile robot indoor localization using SURF algorithm based on LRF sensor
Wang, Z., Hwang, Y.-S., Kim, Y.-K., et al. (2015) · Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE) · doi:10.1109/sice.2015.7285500 · Cited by 3
Joint Audio-Visual Processing for Video Copy Detection
Liu, Z., Zavesky, E., Gibbon, D. C., & Shahraray, B. (2014) · Academic press library in signal processing · doi:10.1016/b978-0-12-420149-1.00013-2 · Cited by 3
B-SIFT: A Highly Efficient Binary SIFT Descriptor for Invariant Feature Correspondence
Li, J., & Lu, Z. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-31919-8_55 · Cited by 3
A Fast Method for Feature Matching Based on SURF
Jiang, Z., Wang, Q., & Cui, Y. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-31919-8_48 · Cited by 3
PERFORMANCE EVALUATION OF POINT MATCHING METHODS IN VIDEO SEQUENCES WITH ABRUPT MOTIONS
Elloumi, W., Treuillet, S., Leconge, R., et al. (2010) · International Conference on Computer Vision Theory and Applications · doi:10.5220/0002823304270430 · Cited by 3
Real time non-rigid surface detection based on binary robust independent elementary features
Wang, C.-M. (2015) · Journal of Applied Research and Technology · doi:10.1016/j.jart.2015.06.005 · Cited by 3
“In 2012, A. Alahi proposed FAST Retina Keypoint (FREAK) (Alahi et al., 2012) to establish descriptors based on the concept behind the human retina and claimed that the approach could lead to better results than BRISK, SURF and SIFT.”
From text vocabularies to visual vocabularies: what basis?
Martinet, J. (2014) · HAL · Cited by 4
WEAR++: 3D model driven camera tracking on board the International Space Station
Tingdahl, D., Weerdt, D. D., Vergauwen, M., et al. (2011) · International Conference on D Imaging · doi:10.1109/ic3d.2011.6584394 · Cited by 3
“They use SURF features [2] for pose initialisation and Lucas-Kanade optical flow during tracking, arguing that SURF is too slow to be used in real time.”
MethodSelf-citationBerezny, N., de Greef, L., Jensen, B., et al. (2012) · IEEE International Conference on Technologies for Practical Robot Applications · doi:10.1109/tepra.2012.6215654 · Cited by 3
“To avoid relying on contrived landmarks, we used SURF feature-matching as the basis for our approach [13], relying on OpenCV’s implementation of SURF descriptors and approximate nearest-neighbor matching for the basic building blocks of the algorithm.”
MethodRobust pedestrian detection and tracking from a moving vehicle
Tuong, N. X., Müller, T., & Knoll, A. (2011) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.871994 · Cited by 4
A framework for implicit human-centered image tagging inspired by attributed affect
Apostolakis, K. C., & Daras, P. (2013) · The Visual Computer · doi:10.1007/s00371-013-0903-4 · Cited by 3
A Robust Homography Estimation Method Based on Keypoint Consensus and Appearance Similarity
Yan, Q. Y., Xu, Y., & Yang, X. (2012) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2012.74 · Cited by 3
“According to the reasons abo set up our model of keypoint consen e is TI , and the other three racted using SURF, and the tions of inlier location error.”
MethodInfluentialA card-playing humanoid playmate for human behavioral analysis
Kim, M.-G., & Suzuki, K. (2012) · Entertainment Computing · doi:10.1016/j.entcom.2012.03.002 · Cited by 3
Real-Time Simulation of Camera Errors and Their Effect on Some Basic Robotic Vision Algorithms
Hinkenjann, A., Roth, T., Millberg, J., et al. (2013) · doi:10.1109/crv.2013.50 · Cited by 3
“With a triangulation step the depth can be calculated from the disparity map.”
MethodMobile Museum Guidance Using Relational Multi-Image Classification
Bruns, E., & Bimber, O. (2010) · International Conference on Multimedia and Ubiquitous Engineering (MUE) · doi:10.1109/mue.2010.5575082 · Cited by 3
“Due to the increased computational performance of mobile phones, object recognition techniques have shifted away from traditional client-server-based approaches where the phones only act as input/output devices [11], [13], [17], [19], to local on-device object classification systems: Takacs et al. [25] implemented a p…”
BackgroundFusing visual tags and inertial information for indoor navigation
Zachariah, D., & Jansson, M. (2012) · IEEE/ION Position Location and Navigation Symposium · doi:10.1109/plans.2012.6236924 · Cited by 3
Face Recognition using VLAD and its Variants
Vinay, A., Shekhar, V. S., Kumar, C. A., Rao, A. S., Shenoy, G. R., Murthy, K. N. B., & Natarajan, S. (2015) · International Conference on Computer and Communication Technology · doi:10.1145/2818567.2818649 · Cited by 3
“Feature Description is one of the most crucial phases in FR as it governs the overall performance and there are several distinguished algorithms available for carrying out this task such as SIFT (Scale Invariant Face Recognition) [6][20], SURF (Speeded-up Robust Features) [4][5][21], ORB (Oriented Fast Permission to m…”
MethodIntegration of recognition and planning for robot hand grasping
Shin, Y.-D., Jang, G.-R., Park, J., et al. (2013) · International Conference on Ubiquitous Robots and Ambient Intelligence · doi:10.1109/urai.2013.6677505 · Cited by 3
“In this approach, the scale invariant feature transform (SIFT) [3] or speeded up robust features (SURF)[4] algorithm is used with the 2D intensity image for object recognition.”
MethodAutomatic Detection of Object of Interest and Tracking in Active Video
Huang, J., & Li, Z.-N. (2010) · Journal of Signal Processing Systems · doi:10.1007/s11265-010-0540-3 · Cited by 4
Feature matching algorithm based on KAZE and fast approximate nearest neighbor search
Ze-Ping, C., & Xiao, D. (2014) · Advances in intelligent systems research/Advances in Intelligent Systems Research · doi:10.2991/csss-14.2014.63 · Cited by 3
Urban photograph localization using the INSTREET application—accuracy and performance analysis
Grega, M., & Łach, S. (2013) · Multimedia Tools and Applications · doi:10.1007/s11042-013-1538-1 · Cited by 3
3D correspondences in textured depth-maps through planar similarity transform
Marcon, M., Frigerio, E., Sarti, A., et al. (2012) · doi:10.1109/espa.2012.6152434 · Cited by 3
Gong, X., & Liu, J. (2011) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2011.6115602 · Cited by 3
“As the compass operator does, many algorithms take some form of histogramming, for instances, SIFT [6], SURF [7] and geometric blur [8].”
MethodOno, K., & Lee, J. (2013) · IEEE/SICE International Symposium on System Integration (SII) · doi:10.1109/sii.2013.6776754 · Cited by 3
“We adopt SURF (Speed Up Robust Features) [10] as features detection and feature description.”
MethodDoes All Newborn Face Lookalike to Human and Machine?
Tiwari, S., & Singh, S. K. (2015) · Proceedings of the National Academy of Sciences India Section A Physical Sciences · doi:10.1007/s40010-014-0196-7 · Cited by 3
Kaveev, I. N., Ташлинский, А. Г., & Kurbanaliev, R. M. (2011) · Pattern Recognition and Image Analysis · doi:10.1134/s1054661811020477 · Cited by 3
Application of a feature-based approach to debris flow detection by numerical simulation
Lin, C.-W., Chen, C.-W., Chen, C.-W., et al. (2013) · Natural Hazards · doi:10.1007/s11069-013-0605-2 · Cited by 3
Multiple 3D Object Tracking using ROI and Double Filtering for Augmented Reality
Park, J.-H., Seo, D.-H., Ku, M.-S., et al. (2011) · doi:10.1109/mue.2011.64 · Cited by 3
An improved system for concept-based video retrieval
Guo, X., Zhao, Z., Chen, Y., et al. (2012) · IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC) · doi:10.1109/icnidc.2012.6418781 · Cited by 3
“[8], which is based on Harr wavelet responses rather than oriented gradients and is cheaper to compute than the Gaussian derivatives that are used in SIFT.”
MethodInexact matching of structural models based on the duality of patterns and classifiers
Stommel, M., Kuhnert, K.-D., & Xu, W. (2014) · Pattern Analysis and Applications · doi:10.1007/s10044-014-0384-8 · Cited by 3
“This includes for example SIFT [20] and SURF [7] features.”
MethodFast vision-based localization for a mars airplane
Arai, K., Takamura, H., Inoue, H., Ono, M., & Adachi, S. (2014) · SICE Annual Conference (SICE) · doi:10.1109/sice.2014.6935277 · Cited by 3
“There have been many studies in the field of computer vision on the extension methods for SIFT to reduce the computation time, such as PCA-SIFT [7] and the speeded up robust features [8] (SURF) method.”
MethodA robust learning algorithm based on SURF and PSM for facial expression recognition
Chen, J., Lin, X., Takiguchi, T., et al. (2014) · doi:10.1109/icosp.2014.7015220 · Cited by 3
Equivalence of BRISK Descriptors for the Registration of Variable Bit-Depth Aerial Imagery
Griffith, E. J., Yuan, C., Jump, M., et al. (2013) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2013.442 · Cited by 3
“SIFT (Scale-Invariant Feature Transform) a local invariant feature descriptor was published in 1999 [9], followed by a faster and improved form of SIFT known as SURF (SpeededUp Robust Features) in 2006 [10].”
MethodGeometrical 3D reconstruction using real-time RGB-D cameras
Penelle, B., Schenkel, A., & Warzée, N. (2011) · International Conference on D Imaging · doi:10.1109/ic3d.2011.6584368 · Cited by 3
On-line document registering and retrieving system for AR annotation overlay
Uchiyama, H., Pilet, J., & Saitô, H. (2010) · Augmented Human International Conference · doi:10.1145/1785455.1785478 · Cited by 3
Reconfigurable cache memory architecture for integral image and integral histogram applications
Hsu, P. Y.-T., & Chien, S. (2011) · SiPS ... design and implementation - IEEE Workshop on Signal Processing Systems · doi:10.1109/sips.2011.6088966 · Cited by 3
A new approach to map joining for depth-augmented visual SLAM
Liu, C., & Song, K. (2013) · Asian Control Conference (ASCC) · doi:10.1109/ascc.2013.6606243 · Cited by 3
“In feature extraction and matching procedure, we adopted the Speed Up Robust Feature (SURF)[14] to extract features for SLAM calculation.”
MethodInfluentialStitching giga pixel images using parallel computing
Kooper, R., Bajcsy, P., & Hernández, N. M. (2010) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.873645 · Cited by 3
Action recognition of motion capture data
Lv, N., Feng, Z., Ran, L., & Zhao, X. (2014) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp.2014.7003743 · Cited by 3
“The works [15]–[17] use the space-time interest point detectors.”
BackgroundAdaptive feature descriptor selection based on a multi-table reinforcement learning strategy
Piñol, M., Sappa, A. D., & Toledo, R. (2014) · Neurocomputing · doi:10.1016/j.neucom.2014.03.080 · Cited by 3
Mobile Visual Recognition on Smartphones
Gui, Z., Wang, Y., Liu, Y., et al. (2013) · Journal of Sensors · doi:10.1155/2013/843727 · Cited by 3
FineMesh: High-Density Sampling Platform Using an Autonomous Robot
Kawauchi, K., & Rekimoto, J. (2012) · IEEE International Conference on Green Computing and Communications · doi:10.1109/greencom.2012.56 · Cited by 3
1-point affine RANSAC for scene image matching in appearance-based localization
Chou, C. C., & Wang, C. (2014) · 2022 IEEE 18th International Conference on Automation Science and Engineering (CASE) · doi:10.1109/coase.2014.6899478 · Cited by 3
Scene Understanding: A Framework for Image Segmentation via Object Recognition
Loke, K. S., & Egerton, S. J. (2010) · International Conference on Intelligent Environments (IE) · doi:10.1109/ie.2010.67 · Cited by 3
Point cloud registration using MSSIR: Maximally stable shape index regions
Manafzade, M. M., & Harati, A. (2013) · Iranian Conference on Electrical Engineering · doi:10.1109/iraniancee.2013.6599620 · Cited by 3
“Some such methods are extended techniques that have already been applied in RGB and gray images like SIFT [3, 4] and SURF [5, 6].”
MethodMore effective image matching with scale invariant feature transform
Ancuți, C., & Bekaert, P. (2007) · Spring Conference on Computer Graphics · doi:10.1145/2614348.2614359 · Cited by 3
“Speed Up Robust Features (SURF) [Bay et al. 2006] uses Hessian matrix and relies on the integral images to increase the computation ef.ciency for extracting feature points.”
MethodFeature extraction based on hierarchical growing neural gas for informationally structured space
Toda, Y., & Kubota, N. (2013) · 2022 International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2013.6706825 · Cited by 3
Sampaio, J. C., Evangelista, R. D. S., & Fernandes, L. A. F. (2015) · ACM Symposium on Applied Computing · doi:10.1145/2695664.2695810 · Cited by 3
“SURF [4] out-performs SIFT in computational speed.”
Illumination modelling and optimization for indoor video surveillance
Konda, K. R., & Conci, N. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2039884 · Cited by 3
News story clustering from both what and how aspects
Chu, W.-T., Huang, C.-C., & Cheng, W. (2011) · doi:10.1145/2072552.2072555 · Cited by 3
“From each selected frame we extract the SURF (Speech Up Robust Features) feature points [4].”
MethodRobust template matching based obstacle tracking for autonomous rovers
Mishra, P., Kishore, J. K., Shetty, R., et al. (2013) · IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT) · doi:10.1109/conecct.2013.6469290 · Cited by 3
“Moreover, techniques such as Speeded up Robust Features (SURF) [6] and Scale Invariant feature transform (SIFT) [7] in combination with Harris detector [8] have been proposed to identify and track certain features in the image.”
MethodAdvanced feature point transformation of corner points for mobile object recognition
Yin, X., Kim, D.-H., Hong, C.-P., et al. (2014) · Multimedia Tools and Applications · doi:10.1007/s11042-014-2241-6 · Cited by 3
Dawood, M., Najjar, M. E. E., Cappelle, C., et al. (2012) · IEEE International Conference on Vehicular Electronics and Safety (ICVES) · doi:10.1109/icves.2012.6294287 · Cited by 3
Sitting Posture Recognition and Location Estimation for Human-Aware Environment
Manabe, Y., & Sugawara, K. (2011) · International Journal of Software Science and Computational Intelligence · doi:10.4018/jssci.2011010103 · Cited by 3
Image features-based mobile robot localization
Lin, R., Wang, Z., Sun, R., & Sun, L. (2012) · IEEE International Conference on Information and Automation (ICIA) · doi:10.1109/icinfa.2012.6246823 · Cited by 3
Lee, C., Park, J., Bahn, W., Kim, T.-I., Lee, T.-J., Shaikh, M. M., Kim, K., & Cho, D. (2011) · Annual Conference of the IEEE Industrial Electronics Society (IECON) · doi:10.1109/iecon.2011.6119273 · Cited by 3
“[3] H. Bay, T. Tuytelaars, and Gool, “Surf: Speeded up robust features,” Computer Vision - ECCV 2006, vol.3951, pp. 404-417.”
MethodA Novel High Accuracy Sub-Pixel Corner Detection Algorithm for Camera Calibration
Yu, F. J., Luan, X., Song, D. L., et al. (2012) · Applied Mechanics and Materials · doi:10.4028/www.scientific.net/amm.239-240.713 · Cited by 3
“Tissainayagam and Suter D. [10] compared different corner detectors, and proved that Harris method has better performance than SURF [11] and SUSAN [12] in balance between accuracy and real-time.”
MethodUmetani, T., Yamane, S., & Tamura, Y. (2012) · International Conference on Ubiquitous Robots and Ambient Intelligence · doi:10.1109/urai.2012.6462985 · Cited by 3
Real-Time and Robust Stereo Visual Navigation Localization Algorithm Based on ORB
Xu, Y. X., & Chen, F. (2012) · Applied Mechanics and Materials · doi:10.4028/www.scientific.net/amm.241-244.478 · Cited by 3
Kim, S., Kwon, S. C., & Kweon, I. S. (2012) · Cognitive Computation · doi:10.1007/s12559-012-9194-8 · Cited by 3
Ballesta, M., Gil, A., Reinoso, Ó., et al. (2010) · Revista Iberoamericana de Automática e Informática Industrial RIAI · doi:10.1016/s1697-7912(10)70027-3 · Cited by 3
“La versión U-SURF no es invariante a rotación y el vector tiene 64 elementos (Bay et al. (2006)).”
BackgroundMulti-sensor image registration using edge-enhanced maximally stable extremal region
Liu, L., Tuo, H., Xu, T., et al. (2012) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp.2012.6469944 · Cited by 3
Recognition of the objects on the conveyor belt using graph matching algorithms
Ondroušek, V., Kolomazník, J., & Vytečka, M. (2014) · International Conference on Mechatronics - Mechatronika · doi:10.1109/mechatronika.2014.7018350 · Cited by 3
“Three different approaches to solving this issue were implemented in our solution, the SIFT [2], the SURF [3] and the approach based on Canny edge detector [4].”
MethodCC-SIFT: Exploiting chromatic contrast forwide-baseline matching
Ancuti, C. O., Ancuți, C., & Bekaert, P. (2010) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2010.5495289 · Cited by 3
Accurate stereo 3D point cloud generation suitable for multi-view stereo reconstruction
Kordelas, G., Daras, P., Klavdianos, P. B. D. L., et al. (2014) · doi:10.1109/vcip.2014.7051565 · Cited by 3
Object tracking based on local feature points
Wang, H., & Zhang, L. (2010) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp.2010.5648034 · Cited by 3
“We present a method of combining RANSAC [11] and motion location correlation information with SURF [12] [13] features through consecutive frames to discard false match.”
MethodStel Component Analysis: Joint Segmentation, Modeling and Recognition of Objects Classes
Perina, A., Jojić, N., Cristani, M., & Murino, V. (2012) · International Journal of Computer Vision · doi:10.1007/s11263-012-0536-5 · Cited by 3
Robust human action recognition using improved BOW and hybrid features
Vo, V., & Ly, N. Q. (2012) · IEEE International Symposium on Signal Processing and Information Technology (ISSPIT) · doi:10.1109/isspit.2012.6621291 · Cited by 3
Speeded-up robust features (SURF) as a benchmark for heterogeneous computers
Iparraguirre, J., Balmaceda, L., & Mariani, C. (2014) · IEEE Biennial Congress of Argentina (ARGENCON) · doi:10.1109/argencon.2014.6868545 · Cited by 3
“CLSurf does the necessary math to obtain the detector and the descriptor (including the determinant of the Hessian and the Haar transform [18]) in kernels that runs on GPU.”
Background
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.