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 6,601–6,700 of 14,752
CoMatcher: Multi-View Collaborative Feature Matching
Zhang, J., Xia, Z., Dong, M., Shen, S., Yue, L., & Zheng, X. (2025) · 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/cvpr52734.2025.02046 · Cited by 16
“Modern methods leverage deep networks for representation learning to extract feature points [10, 13, 14, 39, 53, 59], outperforming traditional hand-crafted encodings [5, 31, 41] and achieving significant advancements on mainstream benchmarks [3, 25, 46].”
Matrix3D: Large Photogrammetry Model All-in-One
Lu, Y., Zhang, J., Fang, T., Nahmias, J.-D., Tsin, Y., Quan, L., Cao, X., Yao, Y., & Li, S. (2025) · 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/cvpr52734.2025.01051 · Cited by 51
“These pipelines [17, 18, 35, 87] typically begin with camera parameter estimation through feature matching [5, 34, 63, 83] across images, followed by bundle adjustment to jointly optimize the 3D point cloud and camera poses.”
Manjunath, D., Sikdar, A., Gurunath, P., Udupa, S., & Sundaram, S. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00443 · Cited by 15
“RGB camera 1280x720 400-700nm 60”
DTM generation from STC-SIMBIO-SYS images
Re, C., Simioni, E. T., Cremonese, G., et al. (2015) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2184745 · Cited by 6
SERP: SURF Enhancer for Repeated Pattern
Mok, S. J., Jung, K., Ko, D. W., Lee, S. H., & Choi, B.-U. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-24031-7_58 · Cited by 5
Scalable Top-$$ k$$ Spatial Image Search on Road Networks
Zhao, P., Kuang, X., Sheng, V. S., Xu, J., Wu, J., & Cui, Z. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-18123-3_23 · Cited by 5
CWMA: Circular Window Matching Algorithm
Miramontes-Jaramillo, D., Kober, V. I., & Diaz-Ramirez, V. H. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-41822-8_55 · Cited by 6
Image Subset Selection Using Gabor Filters and Neural Networks
Ali, H. K., & Whitehead, A. D. (2015) · The International journal of Multimedia & Its Applications · doi:10.5121/ijma.2015.7204 · Cited by 6
Zhang, H., & Li, X. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-40099-5_10 · Cited by 6
A robust and lightweight feature system for video fingerprinting
Liu, T.-J., Han, H. J., Xin, X., Li, Z., & Katsaggelos, A. K. (2012) · Cited by 6
Identifying an efficient and robust sub-shot segmentation method for home movie summarisation
Cooray, S. H., & O'Connor, N. E. (2010) · International Conference on Intelligent Systems Design and Applications (ISDA) · doi:10.1109/isda.2010.5687086 · Cited by 6
“We identified that a local feature tracker like the PLK algorithm performs better than the SIFT and SURF algorithms on sub-shot detection in home movie content.”
MethodInfluentialLarge-scale visual concept detection with explicit kernel maps and power mean SVM
Sjöberg, M., Koskela, M., Ishikawa, S., et al. (2013) · doi:10.1145/2461466.2461505 · Cited by 6
“Current state-of-the-art automatic concept detectors rely on local features such as SIFT [16] or SURF [1], bag-ofvisual-words (BoV) histograms generated typically by kmeans (or recently also with kernel k-means [38] or with GMMs), and computationally heavy kernel-based classifiers, such as non-linear SVMs, trained wit…”
MethodClustering based binary descriptor coding for efficient transmission in visual sensor networks
Monteiro, P., & Ascenso, J. (2013) · Picture Coding Symposium · doi:10.1109/pcs.2013.6737674 · Cited by 6
“However, while SIFT describe the area (patch) around each keypoint with an orientation histogram from the smoothed samples for each region, SURF employs Haar wavelet filters computed over a predefined set of sample points around each keypoint.”
BackgroundMethodInfluential3D reconstruction for sinusoidal motion based on different feature detection algorithms
Zhang, P., Zhang, J., Deng, H., et al. (2015) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2181388 · Cited by 6
A motion-information-based vision-tracking system with a stereo camera on mobile robots
Bahn, W., Park, J., Lee, C., et al. (2011) · doi:10.1109/ramech.2011.6070491 · Cited by 6
GPU-Based Computation of the Integral Image
Huang, W., Wu, L., & You-gen, Z. (2011) · International Conference on Virtual Reality and Visualization · doi:10.1109/icvrv.2011.43 · Cited by 6
“In the SURF algorithm [3][4], it speeds up the computation of the first and second order of Gaussian derivative that uses the integral image.”
MethodSequential scene parsing using range and intensity information
Brucker, M., Léonard, S., Bodenmüller, T., et al. (2012) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2012.6224978 · Cited by 6
“Algorithm 1 AdjustExistingObject function ADJUSTEXISTINGOBJECT(Q,I ,o (,tmin,tsuf )) F = EXTRACTSURF(I,Q) P = CALCULATE3DPOSITIONS(I ,F ) 〈Fo, Po〉 = GETFEATURESAND3DPOSITIONS(o) M = FLANN(F ,Fo) 〈µ, n〉 = RANSAC(P ,Po,M ,tsuf ) if n tmin then return 〈∅, n〉 else return 〈µ, n〉 end if end function In order to estimate the…”
MethodInfluentialCompounding Local Invariant Features and Global Deformable Geometry for Medical Image Registration
Zhang, J., Chen, L., Wang, X., Teng, Z., Brown, A. J., Gillard, J. H., Guan, Q., & Chen, S. (2014) · PLoS ONE · doi:10.1371/journal.pone.0105815 · Cited by 6
“Therefore the proposed model can be easily integrated with other local invariant features, such as the maximally stable extremal regions (MSER) [33] and the speeded up robust features (SURF) [34].”
MethodOn accuracy, robustness, and security of bag-of-word search systems
Voloshynovskiy, S., Diephuis, M., Kostadinov, D., Farhadzadeh, F., & Holotyak, T. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2038149 · Cited by 8
“i=1 p(y|x(m)), (1) which reduces to p(y|x(m)) ≡ ∏J′ k=1 p(y |x(m)) in the synchronised case, i.”
Image compression and retrieval for Mobile Visual Search
Cao, Y., Ritz, C. H., & Raad, R. (2012) · International Symposium on Communications and Information Technologies · doi:10.1109/iscit.2012.6380842 · Cited by 6
Ensuring Quality of Experience for markerless image recognition applied to print media content
Davis, S., Cheng, E., Ritz, C. H., et al. (2012) · International Workshop on Quality of Multimedia Experience (QoMEX) · doi:10.1109/qomex.2012.6263857 · Cited by 6
“Of the popular keypoint and descriptor alg this paper, SIFT [4] extracts local image feature to scale, orientation and partially invariant to il SURF [5] descriptors are based on SIFT, and rotation-invariant, the algorithm is much les complex.”
BackgroundMethodSPARTAN project: On profiling computer vision algorithms for rover navigation
Diamantopoulos, D., Siozios, K., Lentaris, G., Soudris, D., & Rodrigalvarez, M. A. (2012) · NASA/ESA Conference on Adaptive Hardware and Systems (AHS) · doi:10.1109/ahs.2012.6268647 · Cited by 9
Classification of Urban Scenes from Geo-referenced Images in Urban Street-View Context
Iovan, C., Picard, D., Thome, N., et al. (2012) · International Conference on Machine Learning and Applications · doi:10.1109/icmla.2012.171 · Cited by 6
A CMOS-3D reconfigurable architecture with in-pixel processing for feature detectors
Suarez, M., Brea, V. M., Pardo, F., et al. (2012) · doi:10.1109/3dic.2012.6263019 · Cited by 6
“Feature detectors as Harris [2], Harris-affine [3], Hessian [4], Hessian-affine [5], Scale Invariant Feature Transform (SIFT) [1] or Speeded Up Robust Feature (SURF) [6] identify certain pixels in the image and make a correspondence with pixels in other image to perform the corresponding task.”
MethodCamera calibration in wireless multimedia sensor networks
Liu, J., O’Rourke, D., Wark, T., et al. (2009) · International Conference on Intelligent Sensors, Sensor Networks and Information Processing · doi:10.1109/issnip.2009.5416797 · Cited by 6
“SURF feature extractor and descriptor [1] is employed, due to its robustness and speed, to find corresponding points between the images.”
BackgroundMethodExtended 3D Line Segments from RGB-D Data for Pose Estimation
Buch, A. G., Jessen, J. B., Kraft, D., et al. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-38886-6_6 · Cited by 6
“SIFT and SURF use their own keypoint detectors for locating interest points in the images.”
MethodContour-based hidden Markov model to segment 2D ultrasound images
Qian, X., & Yoon, B.-J. (2011) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2011.5946501 · Cited by 7
Sun, B., Xu, F., Zhou, G., et al. (2015) · Journal of Electronic Imaging · doi:10.1117/1.jei.24.1.013018 · Cited by 6
“To address the alignment problem in SRC, some methods 22,23,24 introduced the scale-invariant feature transform (SIFT) 22 or the speeded-up robust features 25 descriptor to the recognition method.”
MethodMotor-Primed Visual Attention for Humanoid Robots
Lukic, L., Billard, A., & Santos-Victor, J. (2015) · IEEE Transactions on Autonomous Mental Development · doi:10.1109/tamd.2015.2417353 · Cited by 7
Occlusion-Aware Motion Layer Extraction Under Large Interframe Motions
Xu, F., & Dai, Q. (2011) · IEEE Transactions on Image Processing · doi:10.1109/tip.2011.2121081 · Cited by 6
A semantically significant visual representation for social image retrieval
Sayad, I. A. E., Martinet, J., Urruty, T., et al. (2011) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2011.6011867 · Cited by 6
“We utilize the SURF [9] descriptor with the Edge context descriptor [10].”
MethodTowards condition-invariant, top-down visual place recognition
Milford, M. J., Vig, E., Scheirer, W. J., & Cox, D. (2013) · QUT ePrints (Queensland University of Technology) · Cited by 6
Object detection using voting spaces trained by few samples
Xu, P., Ye, M., Li, X., Pei, L., & Jiao, P. (2013) · Optical Engineering · doi:10.1117/1.oe.52.9.093105 · Cited by 6
“F1ðQ;HiÞ 1⁄4 1 pðHijLQÞ ≥ τL; 0 otherwise; (18)”
Fast and low complexity image stitching method on mobile phones
Yan, W., Liu, C., & Luo, W. (2015) · International Conference on Control, Automation and Information Sciences (ICCAIS) · doi:10.1109/iccais.2015.7338670 · Cited by 6
AUTOMATIC GEOREFERENCING OF AERIAL IMAGES USING HIGH- RESOLUTION STEREO SATELLITE IMAGES
Oh, J., Toth, C. K., & Grejner‐Brzezinska, D. A. (2010) · Cited by 6
Large-scale high-dimensional nearest neighbor search using flash memory with in-store processing
Jun, S.-W., Chung, C., & Arvind, A. (2015) · International Conference on ReConFigurable Computing and FPGAs (ReConFig) · doi:10.1109/reconfig.2015.7393324 · Cited by 10
“In the worst case, a query has to iteratively compute the distance function between the query point and all points in the dataset to determine the nearest neighbor.”
BackgroundHierarchical and binary spatial descriptors for lung nodule image retrieval
Ng, G. M., Song, Y., Cai, W., et al. (2014) · doi:10.1109/embc.2014.6945108 · Cited by 6
“We will also perform more comprehensive evaluation by comparing with the other more recent feature descriptors, such as SURF.”
BackgroundMulti-view distributed source coding of binary features for visual sensor networks
Monteiro, N., Brites, C., Pereira, F. M. B., et al. (2016) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2016.7472189 · Cited by 6
Semantic-aware framework for Mobile Image Search
Bouhlel, N., Ksibi, A., Ammar, A. B., & Amar, C. B. (2015) · International Conference on Intelligent Systems Design and Applications (ISDA) · doi:10.1109/isda.2015.7489162 · Cited by 6
Facilitating Monkey Test by Detecting Operable Regions in Rendered GUI of Mobile Game Apps
Sun, C., Zhang, Z., Jiang, B., & Chan, W. K. (2016) · IEEE International Conference on Software Quality, Reliability and Security (QRS) · doi:10.1109/qrs.2016.41 · Cited by 6
A novel digital watermark against RST distortion based on SURF
Zhang, B., Wang, J., Wen, J., et al. (2010) · IEEE International Conference on Information Theory and Information Security · doi:10.1109/icitis.2010.5688742 · Cited by 6
“Bay [1] proposed an approximation to the Laplacian of Gaussians by using box filter representations of the respective kernels.”
MethodOn road vehicle make and model recognition via sparse feature coding
Nazemi, A., Shafiee, M. J., & Azimifar, Z. (2013) · Iranian Conference on Machine Vision and Image Processing · doi:10.1109/iranianmvip.2013.6780025 · Cited by 6
Zhang, X., Ma, S., Wang, S., et al. (2016) · DCC · doi:10.1109/dcc.2016.18 · Cited by 6
Human-Centered Video Feature Selection via mRMR-SCMMCCA for Preference Extraction
Ogawa, T., YAMAGUCHI, Y., Asamizu, S., et al. (2017) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2016edl8126 · Cited by 6
F-SORT: An Alternative for Faster Geometric Verification
Chan, J., Lee, J. A., & Kemao, Q. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-54181-5_25 · Cited by 6
Learning moment-based fast local binary descriptor
Bellarbi, A., Zenati, N., Otmane, S., & Belghit, H. (2017) · Journal of Electronic Imaging · doi:10.1117/1.jei.26.2.023006 · Cited by 6
“The main contribution of SURF is that it overcame SIFT ’ s main weakness, its computational complexity and the resulting low execution speed.”
InfluentialReliable Indoor Navigation on Humanoid Robots using Vision-Based Localization
Moulard, T., Alcantarilla, P. F., Lamiraux, F., et al. (2012) · Cited by 6
Kuang, Y., Åström, K., Kopp, L., Oskarsson, M., & Byröd, M. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-19282-1 · Cited by 7
Improving point matching on multimodal images using distance and orientation automatic filtering
Anzid, H., Goïc, G. L., Bekkarri, A., et al. (2016) · doi:10.1109/aiccsa.2016.7945753 · Cited by 6
“The registration method in this paper is based on SURF method which allows a fast extraction of robust keypoints and ensures detection of enough shared keypoint in different locations for matching step with respect to repeatability and a fast computational time [20].”
MethodLeft ventricle motion estimation for cardiac cine MRI using graph matching
Guo, W., Yang, X., Wu, J., & Lin, A. (2017) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm.2017.8217743 · Cited by 6
Shafiee, M. J., Chwyl, B., Li, F., Chen, R., Karg, M. E., Scharfenberger, C., & Wong, A. M. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.03.028 · Cited by 8
“Scale-invariant feature transform (SIFT) [4] or speeded up robust features (SURF) [5] have been applied in a wide variety of applications such as image matching [6], object classification [7], image registration [8] and action recognition [9].”
BackgroundMulti-Level of Feature Extraction and Classification for X-Ray Medical Image
Abdulrazzaq, M. M., Yaseen, I. F. T., Noah, S. A. M., et al. (2018) · Indonesian Journal of Electrical Engineering and Computer Science · doi:10.11591/ijeecs.v10.i1.pp154-167 · Cited by 6
CoMo: a scale and rotation invariant compact composite moment-based descriptor for image retrieval
Vassou, S. A., Anagnostopoulos, N., Christodoulou, K., Amanatiadis, A. A., & Chatzichristofis, S. A. (2018) · Multimedia Tools and Applications · doi:10.1007/s11042-018-5854-3 · Cited by 6
SLAM and Vision-based Humanoid Navigation
Stasse, O. (2018) · Springer eBooks · doi:10.1007/978-94-007-6046-2_59 · Cited by 7
Loop-Closure Detection in Urban Scenes for Autonomous Robot Navigation
Maffra, F., Teixeira, L., Chen, Z., et al. (2017) · Repository for Publications and Research Data (ETH Zurich) · doi:10.3929/ethz-b-000235574 · Cited by 6
An adaptive localization system for image storage and localization latency requirements
Mao, J., Hu, X., & Milford, M. J. (2018) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2018.06.007 · Cited by 6
“Local features, such as SIFT [18], SURF [19] and BRISK [20], describe image regions of interest and are widely used in visual place recognition [21-24].”
MethodMoayed, Z., Griffin, A., & Klette, R. (2017) · International Conference on Image and Vision Computing New Zealand (IVCNZ) · doi:10.1109/ivcnz.2017.8402465 · Cited by 6
Real-Time Pen Input System for Writing Utilizing Stereo Vision
Imad, F. (2018) · Journal of Computers · doi:10.17706/jcp.13.9.1000-1009 · Cited by 6
“A more robust tracking algorithm such as SIFT or SURF that doesn’t require a marker can be used to track the pen tip [15], [16].”
MethodJoint Image Restoration and Matching Based on Distance-Weighted Sparse Representation
Shao, Y., Sang, N., Gao, C., & Lin, W. (2018) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2018.8545413 · Cited by 6
“Following are some feature extracting methods: SIFT feature descriptor [2], SURF operator [3], Harris operator [4], SUSAN operator [5], Ridge let transformation [6] and Canny operator [7].”
MethodTowards Automated Biometric Identification of Sea Turtles (Chelonia mydas)
Hipiny, I., Ujir, H. B., Mujahid, A., et al. (2018) · Journal of ICT Research and Applications · doi:10.5614/itbj.ict.res.appl.2018.12.3.4 · Cited by 6
From Neurorobotic Localization to Autonomous Vehicles
Espada, Y., Cuperlier, N., Bresson, G., & Romain, O. (2019) · Unmanned Systems · doi:10.1142/s2301385019410048 · Cited by 7
Modeling of microscope images for early detection of fatigue cracks in structural materials
Ghalyan, N. F., Ghalyan, I. F. J., & Ray, A. (2019) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-019-04108-z · Cited by 6
“Feature extraction of images can be realized by myriad of techniques, such as speeded up robust features (SURF) [4] and scale-invariant feature transform (SIFT) [31].”
MethodStereo Image and Depth Map Generation for Images with Different Views and Resolutions
Kwan, C., Chou, B., & Ayhan, B. (2018) · IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · doi:10.1109/uemcon.2018.8796832 · Cited by 6
Low-latency Visual SLAM with Appearance-Enhanced Local Map Building
Zhao, Y., Ye, W., & Vela, P. A. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794046 · Cited by 6
Ramtoula, B., de Azambuja, R., & Beltrame, G. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9197270 · Cited by 6
Baso, B., & Suciati, N. (2020) · Jurnal Teknologi Informasi dan Ilmu Komputer · doi:10.25126/jtiik.2020722002 · Cited by 6
Duong, N.-D., Soladié, C., Kacete, A., Richard, P.-Y., & Royan, J. (2019) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2019.102850 · Cited by 8
An Ultrahigh-Speed Object Detection Method With Projection-Based Position Compensation
Li, J., Long, X., Xu, D., Gu, Q., & Ishii, I. (2019) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2019.2953418 · Cited by 6
“Other robust image feature descriptors, such as scale-invariant feature transform (SIFT) [4], speeded-up robust features (SURF) [5], binary robust-independent elementary features (BRIEF) [6], and oriented FAST and rotated BRIEF (ORB) [7], are also widely used in the fields of image stitching and…”
MethodImage Local Features Description Through Polynomial Approximation
Fawad, F., Rahman, M., Khan, M. J., Asghar, M. A., Amin, Y., Badnava, S., & Mirjavadi, S. S. (2019) · IEEE Access · doi:10.1109/access.2019.2959326 · Cited by 7
“As mentioned in the literature HRI-CSLTP, LIOP, GLOH, SIFT, V-SIFT, and SURF use Euclidean distance-based matching, while L2SSP, DAISY, and DoP use the correlation-based matching technique.”
MethodInfluentialKazerouni, M. F., Saeed, N. T. M., & Kuhnert, K.-D. (2019) · IEEE International Conference on Intelligent Computer Communication and Processing (ICCP) · doi:10.1109/iccp48234.2019.8959784 · Cited by 8
“To identify plant species, feature extraction algorithms, scale-invariant feature transform (SIFT) [25], speeded up robust features (SURF) [26], oriented FAST and rotated Brief (ORB) [27] and neural networks were used.”
MethodAn Image Orientation Estimation Method Based on Gradient Histogram and Point Matching
Cheng, Z., & Chai, S. (2019) · Chinese Automation Congress · doi:10.1109/cac48633.2019.8997224 · Cited by 6
FCM-Based Approach for Locating Visible Video Watermarks
Embaby, A. A., Shalaby, M. A. W., & Elsayed, K. M. A. (2020) · Symmetry · doi:10.3390/sym12030339 · Cited by 6
A Food Recognition and Tracking System for Diabetics in the Middle East
Usman, M., Ahmad, K., & Qaraqe, M. (2019) · IEEE International Smart Cities Conference (ISC2) · doi:10.1109/isc246665.2019.9071759 · Cited by 6
LPD-AE: Latent Space Representation of Large-Scale 3D Point Cloud
Suo, C., Liu, Z., Mo, L., & Liu, Y. (2020) · IEEE Access · doi:10.1109/access.2020.2999727 · Cited by 6
A Comparative Study of Different Feature Descriptors for Video-Based Human Action Recognition
Sadhukhan, S., Mallick, S., Singh, P. K., et al. (2020) · Advances in intelligent systems and computing · doi:10.1007/978-981-15-4288-6_3 · Cited by 6
Bag-of-Visual Words based Improved Image Retrieval Algorithm for Vision Indoor Positioning
Jia, S., Ma, L., Tan, X., et al. (2020) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2020-spring48590.2020.9129412 · Cited by 6
“Speeded-Up Robust Features (SURF) algorithm improves the SIFT algorithm in terms of the calculation speed [9].”
MethodFractional speeded up robust features detector with the Caputo-Fabrizio derivative
Lavín-Delgado, J. E., Solís-Pérez, J. E., Gómez-Aguilar, J. F., et al. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09547-5 · Cited by 6
“An interest point detector and descriptor called Speeded-Up Robust Feature (SURF) was presented in [5].”
MethodAutomated Classification of Acute Rejection from Endomyocardial Biopsies
Giuste, F., Venkatesan, M., Zhao, C., Li, T., Zhu, Y., Deshpande, S. R., & Wang, M. D. (2020) · ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB) · doi:10.1145/3388440.3412430 · Cited by 8
“For each patch, grey level co-occurrence matrix (GLCM) properties, discrete cosine transform (DCT) features, and SURF descriptors were generated [3].”
MethodConditional Probabilistic Relative Visual Localization for Unmanned Aerial Vehicles
Couturier, A., & Akhloufi, M. A. (2020) · IEEE Canadian Conference on Electrical and Computer Engineering (CCECE) · doi:10.1109/ccece47787.2020.9255691 · Cited by 6
Superpixel-Based Local Features for Image Matching
Dong, Y., Fan, D., Ma, Q., & Ji, S. (2021) · IEEE Access · doi:10.1109/access.2021.3052502 · Cited by 7
“For example, speeded up robust features (SURF) [11] uses the box function to approximate the Hessian matrix, and dense articulated real-time tracking (DART) [12] uses the piecewise trigonometric function to approximate the Hessian matrix.”
MethodTwo-Stage vSLAM Loop Closure Detection Based on Sequence Node Matching and Semi-Semantic Autoencoder
Wang, Z., Peng, Z., Guan, Y. L., & Wu, L. (2021) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-020-01302-0 · Cited by 7
Liu, X., Li, J., Pan, J., et al. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-020-10213-z · Cited by 6
Detecting Loop Closure using Enhanced Image for Underwater VINS-Mono
Zhao, H., Zheng, R., Liu, M., & Zhang, S. (2020) · Global Oceans 2020: Singapore – U.S. Gulf Coast · doi:10.1109/ieeeconf38699.2020.9388996 · Cited by 9
INCAME: Interruptible CNN Accelerator for Multirobot Exploration
Yu, J., Xu, Z., Zeng, S., Yu, C., Qiu, J., Shen, Z., Xu, Y., Dai, G., Wang, Y., & Yang, H. (2021) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2021.3072340 · Cited by 9
SUZE: A Hybrid Approach for Multi-fish Detection and Tracking
Bhateja, A., Lall, B., Kalra, P. K., & Chaudhary, K. (2020) · Global Oceans 2020: Singapore – U.S. Gulf Coast · doi:10.1109/ieeeconf38699.2020.9389154 · Cited by 8
“In this paper, we suggest an advanced approach for producing a feature set that is strong enough, by taking in account a combination of SURF (Speeded up robust features) [6] and KAZE [7] keypoints.”
BackgroundMethodA Scale and Rotational Invariant Key-point Detector based on Sparse Coding
Hong-Phuoc, T., & Guan, L. (2021) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3452009 · Cited by 6
DGD-net: Local Descriptor Guided Keypoint Detection Network
Liu, X., Meng, C., Tian, F.-P., & Feng, W. (2021) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme51207.2021.9428406 · Cited by 10
“Traditional methods usually design handcraft feature detection and description for keypoint matching [7, 8, 9].”
MethodA Convolutional Neural Network for Learning Local Feature Descriptors on Multispectral Images
Nunes, C. F. G., & Pádua, F. L. C. (2021) · IEEE Latin America Transactions · doi:10.1109/tla.2022.9661460 · Cited by 6
Evolutionary Rear-Lamp Tracking at Nighttime
Sun, H., Nakane, T., Zhang, N., & Zhang, C. (2021) · IEEE Access · doi:10.1109/access.2021.3087238 · Cited by 7
A Light Field FDL-HSIFT Feature in Scale-Disparity Space
Xiao, Z., Zhang, M., Jin, H., et al. (2021) · Proceedings - International Conference on Image Processing · doi:10.1109/icip42928.2021.9506361 · Cited by 6
Learning 3D medical image keypoint descriptors with the triplet loss
Loiseau--Witon, N., Kéchichian, R., Valette, S., & Bartoli, A. (2021) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-021-02481-3 · Cited by 6
“These descriptors are essential andmust be discriminant and repeatable [1,2].”
BackgroundMethodDense Robust 3D Reconstruction and Measurement for 3D Printing Process Based on Vision
Lv, N., Wang, C., Qiao, Y., et al. (2021) · Applied Sciences · doi:10.3390/app11177961 · Cited by 6
Zheng, X., Tao, Y., Zhang, R., Yang, W., & Liao, Q. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.09.001 · Cited by 6
An Improved ORB Algorithm for Feature Extraction and Homogenization Algorithm
Bei, Q., Liu, H., Pei, Y., et al. (2021) · IEEE International Conference on Electronic Technology, Communication and Information (ICETCI) · doi:10.1109/icetci53161.2021.9563421 · Cited by 6
SeqMatchNet: Contrastive Learning with Sequence Matching for Place Recognition & Relocalization
Garg, S., Vankadari, M. B., & Milford, M. J. (2021) · Conference on Robot Learning · Cited by 43
Removal of self co-articulation and recognition of dynamic hand gestures using deep architectures
Kirupakaran, A. M., Yadav, K. S., Misra, S., et al. (2021) · Applied Soft Computing · doi:10.1016/j.asoc.2021.108122 · Cited by 6
A Visual Inertial Odometry Framework for 3D Points, Lines and Planes
Kannapiran, S., van Baar, J., & Berman, S. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9636526 · Cited by 6
“To ensure greater robustness to mismatches in 2D feature points, SLAM with points and planes [2], [9] rely on both SIFT [10] (or alternatively SURF [11]) features and plane correspondences.”
Towards Real-time 3D Visualization with Multiview RGB Camera Array
Ke, J., Watras, A. J., Kim, J.-J., Liu, H., Jiang, H., & Hu, Y. H. (2022) · Journal of Signal Processing Systems · doi:10.1007/s11265-021-01729-0 · Cited by 6
“RT3DV is implemented in C++ using OpenCV with CUDA 9 enabled, where SURF [11] and FLANN [39] matching run in GPU.”
MethodBaruah, A. C., & Sychterz, A. C. (2022) · Frontiers in Built Environment · doi:10.3389/fbuil.2022.813543 · Cited by 7
Research on image stitching method based on fuzzy inference
Lu, J., Huo, G., & Cheng, J. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-12748-9 · Cited by 6
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.