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 6,201–6,300 of 14,752
Multi-waypoint visual homing in piecewise linear trajectory
Fu, Y., Hsiang, T.-R., & Chung, S. (2012) · Robotica · doi:10.1017/s0263574712000434 · Cited by 7
“Some common features are SIFT (Scale-Invariant Feature Transform), 34 SURF (Speeded Up Robust Feature), 35 corners, and lines.”
MethodMukhopadhyay, A., Qian, Z. Y., Bhandarkar, S. M., Liu, T., Vörös, S., & Rinehart, S. J. (2014) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2014.2357472 · Cited by 8
Motion dense sampling and component clustering for action recognition
Aihara, K., & Aoki, T. (2014) · Multimedia Tools and Applications · doi:10.1007/s11042-014-2112-1 · Cited by 7
CLSIFT: An Optimization Study of the Scale Invariance Feature Transform on GPUs
Wang, W., Zhang, Y., Guoping, L., et al. (2013) · doi:10.1109/hpcc.and.euc.2013.23 · Cited by 7
Context-based understanding of interaction intentions
Quintas, J. M., Almeida, L., Brito, M., et al. (2012) · doi:10.1109/roman.2012.6343803 · Cited by 7
The Gixel array descriptor (GAD) for multimodal image matching
Pang, G., & Neumann, U. (2013) · Proceedings · doi:10.1109/wacv.2013.6475060 · Cited by 7
Vision-based scale-adaptive vehicle detection and tracking for intelligent traffic monitoring
Elkerdawy, S., Salaheldin, A. M., & Elhelw, M. (2014) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio.2014.7090470 · Cited by 8
Content-based copy detection by a subspace learning based video fingerprinting scheme
Cirakman, O., Günsel, B., Şengör, N. S., & Kutluk, S. (2012) · Multimedia Tools and Applications · doi:10.1007/s11042-012-1269-8 · Cited by 7
Visual odometry based on a Bernoulli filter
Zhang, F., Clarke, D., & Knoll, A. (2015) · International Journal of Control Automation and Systems · doi:10.1007/s12555-014-0192-3 · Cited by 7
“Various features are investigated based on their properties, e. g. Harris corner feature [19], SIFT feature [20], SURF feature and Kanade-Lucas-Tomasi feature [21,22].”
BackgroundStudy on the Side-scan Sonar Image Matching Navigation Based on SURF
Tao, W., Zhao, J., Liu, J., et al. (2010) · International Conference on Electrical and Control Engineering · doi:10.1109/icece.2010.537 · Cited by 7
“If g(σ) is zero mean Gaussian function, which has variance σ, the image pixel of seabed relief in navigation area is m=(x, y), and the image scale is s, then the Hessian matrix([11]) of pixel m can be calculated by (1):”
MethodEye-in-hand/eye-to-hand configuration for a WMRA control based on visual servoing
Elarbi-Boudihir, M., & Al-Shalfan, K. A. (2013) · IEEE International Workshop of Electronics, Control, Measurement, Signals and their Application to Mechatronics (ECMSM) · doi:10.1109/ecmsm.2013.6648971 · Cited by 7
“in 2006 [13], that can be used in computer vision tasks like object recognition or 3D reconstruction.”
BackgroundFast calculating feature point's main orientation in SURF algorithm
Han, B., Wang, Y., & Jia, X. (2010) · International Conference on Computer, Mechatronics, Control and Electronic Engineering · doi:10.1109/cmce.2010.5609874 · Cited by 7
Parallel Integral Image Generation Algorithm on Multi-core System
Wu, Y., Wu, Y.-T., Cho, C.-Y., et al. (2011) · IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA) · doi:10.1109/ispa.2011.45 · Cited by 7
“Since then, it was widely used in the fields of pattern recognition and surveillance system especially for the fast calculation of haar-like features [2-4].”
MethodGastroscopic Image Graph: Application to Noninvasive Multitarget Tracking under Gastroscopy
Wang, B., Hu, W., Liu, J., Si, J., & Duan, H. (2014) · Computational and Mathematical Methods in Medicine · doi:10.1155/2014/974038 · Cited by 7
“In order to employ the most suitable estimation method, five common, widely used feature detectors (Shi-Tomasi, SIFT, FAST, SURF, and CenSurE) [23–27] were selected.”
MethodRobust visual tracking using structural region hierarchy and graph matching
Song, Y.-Z., Li, C., Wang, L., Hall, P. M., & Shen, P. (2012) · Neurocomputing · doi:10.1016/j.neucom.2011.11.030 · Cited by 10
“The latter is estimated in terms of maximum likelihood of local SURF feature observations.”
InfluentialPosition-based navigation using multiple homographies
Montijano, E., & Sagüés, C. (2008) · IEEE International Conference on Emerging Technologies and Factory Automation (ETFA) · doi:10.1109/etfa.2008.4638515 · Cited by 7
“The recently appearance of invariant features in the images [7] [1] has given more robustness and better matching, allowing larger displacements between the compared images.”
BackgroundDevelopment of an algorithm for fast corner points detection
Веркеенко, М. С. (2014) · Journal of Computer and Systems Sciences International · doi:10.1134/s1064230714030162 · Cited by 7
Detection and tracking of moving objects in SLAM using vision sensors
Wang, Y.-T., Feng, Y.-C., & Hung, D.-Y. (2011) · IEEE International Instrumentation and Measurement Technology Conference · doi:10.1109/imtc.2011.5944059 · Cited by 7
Wikireality: Augmenting reality with community driven websites
Gray, D. A., Kozintsev, I. V., Wu, Y., et al. (2009) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2009.5202738 · Cited by 7
“One key difference is that we use extended SURF (128-dimensional) [8] features instead of SIFT because they are faster to compute.”
MethodActivity-based synthesized frame generation in 3DTI video
Chen, S., & Nahrstedt, K. (2013) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2013.6607586 · Cited by 7
“We use feature detection [14] and feature matching [15] to mark the different location of the same feature in two frames (Fig.”
MethodA non-parametric scale-based corner detector
Bellavia, F., Tegolo, D., & Valenti, C. F. (2008) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2008.4760976 · Cited by 7
“Recently, an approximated but faster method, namely Speeded Up Robust Features, was described in [2].”
MethodResultMobile Robot Navigation Based on Artificial Landmarks with Machine Vision System
Yudin, D. A., Stepanovich, K. A., & Zalmanovich, M. V. (2013) · Cited by 7
An efficient algorithm for image copy-move forgery detection based on DWT and SVD
Liu, F., & Feng, H. (2014) · International Journal of Security and Its Applications · doi:10.14257/ijsia.2014.8.5.33 · Cited by 7
An augmented reality platform for wearable assisted living systems
Saracchini, R. F. V., & Ortega, C. A. C. (2015) · Applied Computer Science · Cited by 7
Tracking People in Video Sequences by Clustering Feature Motion Paths
Gudyś, A., Rosner, J., Segen, J., Wojciechowski, K. W., & Kulbacki, M. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-11331-9_29 · Cited by 7
Choi, S. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24471-6_11 · Cited by 7
An Empirical Study of Non-Rigid Surface Feature Matching of Human from 3D Video
Doshi, A., Starck, J., & Hilton, A. (2024) · Journal of Virtual Reality and Broadcasting · doi:10.20385/1860-2037/7.2010.3 · Cited by 7
Al-Kaff, A., de la Escalera, A., & Armingol, J. M. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-27340-2_91 · Cited by 7
“SURF: Speed-Up Robust Features [5] is an orientation and scale invariant detector/descriptor inspired by SIFT detector/descriptor.”
MethodA ROS-Based System for an Autonomous Service Robot
Seib, V., Memmesheimer, R., & Paulus, D. W. R. (2016) · Studies in computational intelligence · doi:10.1007/978-3-319-26054-9_9 · Cited by 7
Probabilistic Subpixel Temporal Registration for Facial Expression Analysis
Sarıyanidi, E., Güneş, H., & Cavallaro, A. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16817-3_21 · Cited by 7
“PSTR outperforms FFT as well as RANSAC-based registration with SURF or MSER features.”
MethodInfluentialTools to Perform Local Dense 3D Reconstruction of Shallow Water Seabed
Avanthey, L., Beaudoin, L., Gademer, A., et al. (2016) · Sensors · doi:10.3390/s16050712 · Cited by 7
A spectral filtering based deep learning for detection of logo and stamp
Nandedkar, A. V., Mukherjee, J., & Sural, S. (2015) · National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG) · doi:10.1109/ncvpripg.2015.7490053 · Cited by 7
Deriving image features for autonomous classification from time-series recurrence plots
Schulz, J., Mentges, A., & Zielinski, O. (2016) · Journal of the European Optical Society Rapid Publications · doi:10.1186/s41476-016-0003-y · Cited by 7
“More sophisticated methods investigate contour lines by Fourier descriptors (e.g. [8]), characteristic inflexions (e.g. [9]) or identification of points of interest in scale space (e.g. [10, 11]).”
MethodA modified topological descriptor for forward looking sonar images
dos Santos, M. M., Zaffari, G. B., Duarte, A. C., Fernandes, D. D. A., Drews, P., & Botelho, S. S. D. C. (2016) · OCEANS 2016 - Shanghai · doi:10.1109/oceansap.2016.7485530 · Cited by 7
“The corners are described by descriptors — image feature extractors — such as the Scale Invariant Feature Transform (SIFT) [8], the Speeded Up Robust Features (SURF) [9], the SIFT + Ransac, or the SURF + Ransac, where the Random Sample Consensus (Ransac) method is responsible for estimating which image keypoints follo…”
MethodStructured Feature Similarity with Explicit Feature Map
Kobayashi, T. (2016) · 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/cvpr.2016.136 · Cited by 7
“In addition to SIFT descriptors, we apply the proposed similarity metric to the SURF detector&descriptor (SURF128) [1].”
BackgroundMethodInfluentialImproved image matching method based on ORB
Li, L., Wu, L., & Gao, Y. (2016) · ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD) · doi:10.1109/snpd.2016.7515941 · Cited by 10
“These have several advantages over the gradient descriptors such as SIFT [4], ASIFT [5] and SURF [6], as they are less computation and more compact to store.”
MethodApproximate Matching in ACSM Dissimilarity Measure
Amelio, A. (2016) · Procedia Computer Science · doi:10.1016/j.procs.2016.08.194 · Cited by 7
“…measure with approximation has been compared to exact ACSM dissimilarity measure and to other two methods based on well-known approximate descriptors for image representation 16 , 17 , 18 : (dis)similarity strategy adopting SIFT features 6 and (dis)similarity method employing SURF features 7 .”
MethodFPGA based accelerated orientation calculation in SIFT using luts
Agrawal, K., & Chowdhury, S. R. (2013) · IEEE Asia Pacific Conference on Postgraduate Research in Microelectronics and Electronics (PrimeAsia) · doi:10.1109/primeasia.2013.6731210 · Cited by 7
“Some implementations have attempted to reduce the computational time by sacrificing the quality of features extracted like in the case of Speed-up feature detectors such as SURF [2] and fast SIFT [3] have thus been proposed.”
MethodTowards Monuments’ Holistic Digital Documentation: the Saint Neophytos Enkleistriotis Case Study
Ioannides, M., Ioannidis, C., Enkleistriotis, A. N., et al. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-48496-9_36 · Cited by 7
THE EXAMPLE OF USING INTENSITY ORTHOIMAGES IN TLS DATA REGISTRATION – A CASE STUDY
Markiewicz, J. S. (2017) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xlii-2-w3-467-2017 · Cited by 7
“Another is based on region detection, or the so-called blob detection, such as the Difference of Gaussian detector (DoG; Lindberg, 1993) implemented in the SIFT algorithm (Lowe, 1992) or the Determinant of Hessian (DoH) detector, utilized in the SURF algorithm (Bay et al., 2006).”
MethodVideo stabilisation with total warping variation model
Wu, H., Xiao, L., Shim, H. J., et al. (2017) · IET Image Processing · doi:10.1049/iet-ipr.2016.0645 · Cited by 7
Hand Gesture Recognition Using SIFT Features on Depth Image
Mahmud, H., Hasan, M. K., Tariq, A.-A., et al. (2016) · Advances in Computer-Human Interaction · Cited by 7
A Comparative Study of Descriptors and Detectors in Multispectral Face Recognition
Mamadou, D., Gouton, P., & Adou, K. J. (2016) · International Conference on Signal-Image Technology & Internet-Based Systems (SITIS) · doi:10.1109/sitis.2016.41 · Cited by 7
“In this paper, a study shows a good compromise between detectors and descriptors which are SURF [8], BRISK [16], FAST [6], FREAK [1] and HARRIS [3] representing an alternative to SIFT [5] which had success.”
MethodDeterministic object tracking using Gaussian ringlet and directional edge features
Krieger, E. W., Paheding, S., Aspiras, T. H., & Asari, V. K. (2017) · Optics & Laser Technology · doi:10.1016/j.optlastec.2017.04.011 · Cited by 8
Computation EvaBio: A Tool for Performance Evaluation in Biometrics
Mahier, J., Hemery, B., El-Abed, M., El-Allam, M., Bouhaddaoui, M., & Rosenberger, C. (2011) · HAL · Cited by 6
Human action and event recognition using a novel descriptor based on improved dense trajectories
Mukherjee, S., & Singh, K. K. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-4980-7 · Cited by 6
“We establish the correspondence between the SURF feature points of the consecutive frames by tracking the points using optical flow [30].”
BackgroundMethodInfluentialImproved data visualisation through nonlinear dissimilarity modelling
Rice, I. (2017) · Pattern Recognition · doi:10.1016/j.patcog.2017.07.016 · Cited by 7
An improved SIFT algorithm for infringement retrieval
Sun, G., Wang, C., Ma, B., et al. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-5060-8 · Cited by 7
Mechatronic system for measuring hot-forged automotive parts based on image analysis
Lins, R. G. (2017) · Transactions of the Institute of Measurement and Control · doi:10.1177/0142331217731619 · Cited by 8
“A crucial task in the extraction of geometric characteristics is to locate homologous points in two acquired images using a technique of local feature matching (Bay et al., 2006).”
MethodМоскаленко, В. В., & Коробов, А. Г. (2017) · Radio Electronics Computer Science Control · doi:10.15588/1607-3274-2017-2-7 · Cited by 7
AdaFlock: Adaptive Feature Discovery for Human-in-the-loop Predictive Modeling
Takahama, R., Baba, Y., Shimizu, N., et al. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11509 · Cited by 7
“For instance, in the computer vision area, SIFT (Lowe 2004) and SURF (Bay, Tuytelaars, and Gool 2006) were designed for general object recognition tasks, and different features were proposed for detailed tasks such as face detection (Viola ∗This work has been done while the first author worked at…”
MethodYoon, J., & Kim, D. (2018) · Journal of Real-Time Image Processing · doi:10.1007/s11554-018-0751-6 · Cited by 6
“In this study, we use the SURF descriptor [1] to represent the gradient patterns of each block.”
MethodLocal multi-feature hashing based fast matching for aerial images
Chen, S., Shi, Y., Zhang, Y., Zhao, J., Zhang, C., & Pei, T. (2018) · Information Sciences · doi:10.1016/j.ins.2018.02.036 · Cited by 7
Ouni, A., Urruty, T., & Visani, M. (2018) · Multimedia Tools and Applications · doi:10.1007/s11042-018-5841-8 · Cited by 7
“Many visual descriptors have been created, among them the Scale Invariant Feature Transform (SIFT) [24] and Speeded-up Robust Features (SURF) [5] became two of the most popular descriptors.”
MethodSIPS: Unsupervised Succinct Interest Points.
Cieslewski, T., & Scaramuzza, D. (2018) · Cited by 7
Image matching algorithm of defects on navel orange surface based on compressed sensing
Xie, X., Ge, S., Xie, M., et al. (2018) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-018-0833-0 · Cited by 7
“Afterwards, Bay et al. (2006) proposed SURF algorithm (Xuanmin 2010; Wu et al. 2014) based on SIFT.”
Visual Scene Prediction for Blind People Based on Object Recognition
Jabnoun, H., Benzarti, F., & Amiri, H. (2017) · International Conference on Computer Graphics, Imaging and Visualization (CGIV) · doi:10.1109/cgiv.2017.19 · Cited by 7
“In recent years, the performance of object recognition has been improved thanks to the introduction of more powerful feature representations (SURF [1], SIFT [15] .”
MethodAn Image Retrieval Algorithm Based on Gist and Sift Features.
Xie, B., Qin, J., Xiang, X., et al. (2018) · Cited by 7
Salamah, S. A., & King, R. D. (2018) · doi:10.1109/asar.2018.8480228 · Cited by 7
Sadeghi, B., Jamshidi, K., Vafaei, A., & Monadjemi, S. A. (2018) · The Visual Computer · doi:10.1007/s00371-018-01616-z · Cited by 8
“After geometrical normalization, like some other famous descriptors [8, 26, 30], to make it invariant to illumination”
BackgroundMethodInfluentialRenò, V., Dimauro, G., Labate, G. M., et al. (2018) · IEEE International Workshop on Metrology for the Sea; Learning to Measure Sea Health Parameters (MetroSea) · doi:10.1109/metrosea.2018.8657861 · Cited by 7
“SURF feature detector is a fast interest-point detector that finds and describes images (or image portions) in terms of relevant keypoints (and their neighborhood), using a fast Hessian matrix-based measure for the detector, and a distribution-based descriptor for the neighborhood [23], [24].”
MethodField scene recognition for self-localization of autonomous agricultural vehicle
Morio, Y., Hanada, Y., Sawada, Y., & Murakami, K. (2019) · Engineering in Agriculture Environment and Food · doi:10.1016/j.eaef.2019.03.001 · Cited by 8
Feedback Unilateral Grid-Based Clustering Feature Matching for Remote Sensing Image Registration
Zheng, Z., Zheng, H. Y., Ma, Y., Fan, F., Ju, J., Xu, B., Lin, M., & Cheng, S. (2019) · Remote Sensing · doi:10.3390/rs11121418 · Cited by 7
A Fast UAV Aerial Image Mosaic Method Based on Improved KAZE
Cui, H., Li, Y., & Zhang, K. (2019) · Chinese Automation Congress · doi:10.1109/cac48633.2019.8997399 · Cited by 7
Multi-Scale Object Detection Using Feature Fusion Recalibration Network
Guo, Z., Zhang, W., Liang, Z., Shi, Y., & Huang, Q. (2020) · IEEE Access · doi:10.1109/access.2020.2980737 · Cited by 8
Evaluation of Image Feature Detection and Matching Algorithms
Ou, Y., Cai, Z., Lü, J., Dong, J., & Ling, Y. (2020) · International Conference on Computer and Communication Systems (ICCCS) · doi:10.1109/icccs49078.2020.9118480 · Cited by 9
“SURF (Speed Up Robust Features) algorithm was first proposed by Bay et al [11]in 2006 and improved in 2008[12].”
MethodCertifiably Optimal and Robust Camera Pose Estimation From Points and Lines
Sun, L., & Deng, Z. (2020) · IEEE Access · doi:10.1109/access.2020.3006056 · Cited by 8
“…have been proposed to solve this problem, including the Perspective-n-Point (PnP) methods (e.g. [1]–[4]) based on feature point detectors (e.g. [5], [6]) or descriptors (e.g. [7], [8]), and the Perspective-n-Line (PnL) methods (e.g. [9]–[11]) based on line segment detectors (e.g. [12]–[14]).”
MethodTowards long‐term UAV object tracking via effective feature matching
Zhao, B., Wang, H., Tang, L., et al. (2020) · Electronics Letters · doi:10.1049/el.2020.1096 · Cited by 7
Bülow, H., Mueller, C. A., Chavez, A. G., Buda, F., & Birk, A. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9197453 · Cited by 6
“…geometric information captured by 3D shape features (e.g., [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], respectively see [33], [34] for overviews and comparisons) or visual features (e.g., [35], [36], [37], [38], [39], [40], [41], [42]) based on 2D texture information.”
Gao, Q., Zhou, Y., Li, G., et al. (2020) · IEEE Access · doi:10.1109/access.2020.3029832 · Cited by 7
Challenges Encountered in Building A Fast and Efficient Surveillance System: An Overview
Rama Moorthy, H., Upadhya, V., Holla, V. V., Shetty, S. S., & Tantry, V. (2020) · International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC) · doi:10.1109/i-smac49090.2020.9243563 · Cited by 9
Homography-Based Egomotion Estimation Using Gravity and SIFT Features
Ding, Y., Baráth, D., & Kúkelová, Z. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-69525-5_17 · Cited by 9
Bellazi, K. M., Marino, R., Lanza-Gutiérrez, J. M., & Riesgo, T. (2020) · IEEE Access · doi:10.1109/access.2020.3042699 · Cited by 7
“…to generate the BoF model: Speeded Up Robust Features (SURF), Scale Invariant Feature Transform (SIFT), and Oriented fast and Rotated brief (ORB) [16]–[18]; and four supervised algorithms: Support Vector Machine (SVM), K-Nearest-Neighbor (KNN), Naive Bayes (NB), and Decision Tree (DT) [15].”
MethodKovač, I., & Marák, P. (2020) · Tatra Mountains Mathematical Publications · doi:10.2478/tmmp-2020-0012 · Cited by 8
“For this purpose, we have decided to use the SIFT [16] and SURF [3] algorithms.”
Method3D Modeling and 3D Materialization of Fluid Art that Occurs in Very Short Time
Tosa, N., Pan, Y., Nakatsu, R., et al. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-65736-9_37 · Cited by 7
Automatic Threat Detection in Single, Stereo (Two) and Multi View X-Ray Images
Tuli, A., Bohra, R., Moghe, T., Chaturvedi, N., Mery, D., & Dhiraj (2020) · IEEE India Conference · doi:10.1109/indicon49873.2020.9342253 · Cited by 9
Lee, S. J., Kim, B. H., & Kim, M. Y. (2021) · IEEE Access · doi:10.1109/access.2021.3078623 · Cited by 7
Liu, P., Wang, F., Teodoro, G., & Kong, J. (2021) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi48211.2021.9434114 · Cited by 9
“The resulting landmark set is M− = {(xr, y), (x, yt)} : sr < γ } , where sr is the distance between the matched SURF features [7].”
BackgroundORB-SLAM2S: A Fast ORB-SLAM2 System with Sparse Optical Flow Tracking
Diao, Y., Cen, R., Xue, F., & Su, X. (2021) · International Conference on Advanced Computational Intelligence (ICACI) · doi:10.1109/icaci52617.2021.9435915 · Cited by 7
Scene Classification of Remotely Sensed Images using Ensembled Machine Learning Models
Deepan, P., & Sudha, L. (2021) · Lecture notes in electrical engineering · doi:10.1007/978-981-16-0289-4_39 · Cited by 7
Ground Visibility in Satellite Optical Time Series Based on A Contrario Local Image Matching
von Gioi, R. G., Hessel, C., Dagobert, T., et al. (2021) · Image Processing On Line · doi:10.5201/ipol.2021.342 · Cited by 7
Camera Handoff for Multi-camera Surveillance
Jirafe, A., Jibhe, M., & Satpute, V. R. (2021) · Algorithms for intelligent systems · doi:10.1007/978-981-33-4862-2_29 · Cited by 7
MOD-SLAM:Visual SLAM with Moving Object Detection in Dynamic Environments
Hu, J., Fang, H., Yang, Q., et al. (2021) · Chinese Control Conference (CCC) · doi:10.23919/ccc52363.2021.9549246 · Cited by 7
Automatic Classification of Diabetic Foot Ulcers Using Computer Vision Techniques
López-Cabrera, J. D., Ruiz-González, Y., Díaz-Amador, R., & Taboada-Crispí, A. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-89691-1_29 · Cited by 7
A Stereo SLAM System With Dense Mapping
Zhang, B., & Zhu, D. (2021) · IEEE Access · doi:10.1109/access.2021.3126837 · Cited by 8
“The approach extracts FAST corners [6] and matches them using the ORB descriptor [7] to estimate the motion between successive frames by 3D-2D and 2D-2D.”
MethodDAR‐Net: Dense Attentional Residual Network for Vehicle Detection in Aerial Images
Li, K., & Wang, B. (2021) · Computational Intelligence and Neuroscience · doi:10.1155/2021/6340823 · Cited by 7
“During the feature extraction step, low-level visual descriptors such as SIFT [2], HOG [3], or SURF [4] are used to extract and encode semantic information from each candidate region.”
MethodA streaming hardware architecture for real-time SIFT feature extraction
Sanchez, H. A. L., & George, A. D. (2021) · International Conference on Field-Programmable Technology (FPT) · doi:10.1109/icfpt52863.2021.9609932 · Cited by 7
Domain Adaptation for Viewpoint Estimation with Image Generation
Wu, X., Lu, C., Gu, C., Wu, K., & Zhu, S. (2021) · International Conference on Control, Automation and Information Sciences (ICCAIS) · doi:10.1109/iccais52680.2021.9624602 · Cited by 8
Research on Vision of Intelligent Car Based on Broad Learning System
Liu, X., & Wu, Y. (2022) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2021.3137801 · Cited by 8
A Multiscale Convolutional Registration Network for Defect Inspection on Periodic Lace Surfaces
Ding, X., Lu, B., & Huang, B. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3150581 · Cited by 5
A Unified Framework for Automobile Instrument Detection System
Zhang, J., Liu, Y., Li, Y., & Yu, J. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3147881 · Cited by 8
Simultaneous Localization and Mapping
Stachniss, C. (2015) · Springer eBooks · doi:10.1007/978-3-662-46900-2_49-1 · Cited by 8
Fast and Reliable Vision-Based Navigation for Real Time Kinematic Applications
Amami, M. M. (2022) · International Journal for Research in Applied Science and Engineering Technology · doi:10.22214/ijraset.2022.40395 · Cited by 7
Ni, L., Huo, C., Zhang, X., Wang, P., Zhang, L., Guo, K., & Zhou, Z. (2022) · Remote Sensing · doi:10.3390/rs14040975 · Cited by 9
Liu, Y., Zhao, C., & Wei, Y. (2022) · Applied Sciences · doi:10.3390/app12063007 · Cited by 7
Biswas, R., Roy, S. S., & Biswas, A. (2022) · International Journal of Computers and Applications · doi:10.1080/1206212x.2022.2092937 · Cited by 7
“[20] introduced speed up robust features (SURF) based on integral image convolution to obtain a roughly estimated Hessian matrix.”
MethodWeak6D: Weakly Supervised 6D Pose Estimation With Iterative Annotation Resolver
Mu, F., Huang, R., Shi, K., Li, X., Qiu, J., & Cheng, H. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3190094 · Cited by 8
“methods extract scale invariant points of interest, such as SIFT [21] or SURF [22].”
MethodSmart Real-Time Vehicle Detection and Tracking System Using Road Surveillance Cameras
Alomari, A. H., & Lebdeh, E. A. (2022) · Journal of Transportation Engineering Part A Systems · doi:10.1061/jtepbs.0000728 · Cited by 7
Bruno, P., Spadea, M. F., Scaramuzzino, S., et al. (2022) · Neural Computing and Applications · doi:10.1007/s00521-022-07642-2 · Cited by 7
“Speeded Up Robust Features (SURF) [12], Maximally Stable Extremal Regions (MSER) [13] and Harris corners [14] were used to identify points to be matched.”
MethodStereovision-Based Ego-Motion Estimation for Combine Harvesters
Chen, H., Chen, J., Guan, Z.-H., Li, Y.-M., Cheng, K., & Cui, Z.-H. (2022) · Italian National Conference on Sensors · doi:10.3390/s22176394 · Cited by 8
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.