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 9,001–9,100 of 14,752
Visual odometer system to build feature based maps for mobile robot navigation
Majdik, A., Tamás, L., Popa, M., et al. (2010) · doi:10.1109/med.2010.5547630 · Cited by 3
“SURF was first introduced by [4], the advantage of this algorithm is the computational speed compared with other performing image feature detectors like Scale-Invariant Feature Transform (SIFT) presented by [5], maintaining approximately the same or even better performances regarding: repeatability, distinctiveness an…”
BackgroundMethod3D environmental mapping of mobile robot using a low-cost depth camera
Qian, K., Ma, X., Fang, F., et al. (2013) · IEEE International Conference on Mechatronics and Automation (ICMA) · doi:10.1109/icma.2013.6617969 · Cited by 3
The causal update filter — A novel biologically inspired filter paradigm for appearance-based SLAM
Sünderhauf, N., Neubert, P., & Protzel, P. (2010) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2010.5653221 · Cited by 4
Robust tracking for interactive social video
Wilk, S., Kopf, S., & Effelsberg, W. (2012) · doi:10.1109/wacv.2012.6163022 · Cited by 3
“Speeded Up Robust Feature (SURF) [2] is a fast method for detecting and extracting feature descriptors that are robust to scaling and rotation.”
MethodContent-Based Pseudoscopic View Detection
Lee, J.-H., Jung, C., Kim, C., et al. (2011) · Journal of Signal Processing Systems · doi:10.1007/s11265-011-0608-8 · Cited by 3
“There are several methods which extract feature points in the image such as Harris corner [14], SIFT [10] and SURF (Speed Up Robust Features) [15].”
MethodLow cost human-robot gaze estimation system
Ishac, K., & Rozado, D. (2014) · doi:10.1145/2686612.2686674 · Cited by 3
Robust PCB Image Alignment using SIFT
Kim, J.-C., Cui, X., Park, E.-S., et al. (2010) · Journal of Institute of Control Robotics and Systems · doi:10.5302/j.icros.2010.16.7.695 · Cited by 3
Loop closure detection using local Zernike moment patterns
Sarıyanidi, E., Şencan, O., & Temeltaş, H. (2013) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2008473 · Cited by 3
Visual SLAM Based on Rigid-Body 3D Landmarks
Loncomilla, P., & Ruíz-del-Solar, J. (2011) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-011-9601-5 · Cited by 3
Exploiting local dependencies with spatial-scale space (S-Cube) for near-duplicate retrieval
Cheng, X., Hu, Y., & Chia, L.-T. (2011) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2011.02.003 · Cited by 3
“Of course, other similar invariant local descriptors, such as SPIN [9] and SURF [2] can also achieve comparable results.”
ResultAn Algorithm Based on SURF for Surveillance Video Mosaicing
Guo, K. Y., Ye, S., Jiang, H. M., et al. (2011) · Advanced materials research · doi:10.4028/www.scientific.net/amr.267.746 · Cited by 3
“Document [11] mentioned that we can extracts feature points with SURF method, which based on SIFT algorithm, used integral image for replacing image convolution, and used the mathematical properties of Hessian matrix for testing eigen values.”
MethodMobile visual object identification: from SIFT-BoF-RANSAC to Sketchprint
Voloshynovskiy, S., Diephuis, M., & Holotyak, T. (2015) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2079984 · Cited by 3
Luo, R. C., Huang, C. Y., & Hsiao, C.-C. (2011) · Proceedings of the Annual Conference of the IEEE Industrial Electronics Society · doi:10.1109/iecon.2011.6120005 · Cited by 3
Practical aspects of visual recognition for indoor mobile positioning
Plósz, S., Kertész, Z., Lukovszki, C., et al. (2013) · IEEE International Conference on Cognitive Infocommunications · doi:10.1109/coginfocom.2013.6719304 · Cited by 3
“The SURF (Speeded Up Robust Features) algorithm is similar to SIFT except it uses a so called Fast-Hessian detector.”
MethodUsing augmented reality for rapid prototyping and collaborative design to model 3D buildings
Clifford, R., Clark, A., & Rogozin, M. (2011) · doi:10.1145/2000756.2000774 · Cited by 3
Object-adaptive tracking for AR guidance system
Park, H., Oh, J., Seo, B.-K., et al. (2008) · ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry (VRCAI) · doi:10.1145/1477862.1477868 · Cited by 3
Robust Relocalization and Its Evaluation for Online Environment Map Construction
Kim, S., Coffin, C., & Höllerer, T. (2010) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2010.243 · Cited by 3
SURF-based spatio-temporal history image method for action representation
Ahad, M. A. R., Tan, J. K., Kim, H., et al. (2011) · IEEE International Conference on Industrial Technology (ICIT) · doi:10.1109/icit.2011.5754412 · Cited by 3
“In this paper, we consider the SURF (SpeededUp Robust Features [22-23]) for interest point detection and exploit these STIP features for further analysis.”
MethodImproving Geolocation by Combining GPS with Image Analysis
Pinho, F., Carvalho, A., & Carreira, R. (2014) · Lecture notes in geoinformation and cartography · doi:10.1007/978-3-319-11463-7_15 · Cited by 3
The Scale-Curvature Connection and its Application to Texture Segmentation
Appleboim, E., Hyams, Y., Krakovski, S., et al. (2013) · Theory and Applications of Mathematics & Computer Science · doi:10.62591/tamcs.2013.1.04 · Cited by 3
Srinivasa, N., Bhattacharyya, R., & Grossberg, S. (2008) · IEEE/RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob) · doi:10.1109/biorob.2008.4762776 · Cited by 3
Exploring User Behaviour on Etsy through Dominant Colors
Aryafar, K., Lynch, C., & Attenberg, J. (2014) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2014.256 · Cited by 3
“Various global image features such as color histograms [7], texture values [8] and shape parameters of easily segmentable regions [9] and localized features such as scale invariant feature transform (SIFT) [10], speeded up robust features (SURF) [11] and histogram of oriented gradients (HOG) [12] are proposed for imag…”
BackgroundImproving topological maps for safer and robust navigation
Murillo, A. C., Abad, P., Guerrero, J. J., et al. (2009) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2009.5354262 · Cited by 3
Learning Domain-Specific Feature Descriptors for Document Images
Ramakrishnan, K., & Bart, E. (2012) · IAPR International Workshop on Document Analysis Systems (DAS) · doi:10.1109/das.2012.49 · Cited by 3
“Some of the most popular descriptors today use histograms of responses of these detectors [1], [2], [3], [4], [5].”
BackgroundAn online SLAM-based mosaicking using local maps for ROVs
Ferreira, F., Veruggio, G., Caccia, M., & Bruzzone, G. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5980521 · Cited by 3
Real-time hand gesture feature extraction using depth data
Huang, H., Ju, Z., & Liu, H. (2014) · International Conference on Machine Learning and Cybernetics · doi:10.1109/icmlc.2014.7009118 · Cited by 3
“Next, local shape-related features, such as radial histogram and modified speeded-up robust features (SURF) [16], are extracted from the image.”
MethodLocal image feature matching for object recognition
Sushkov, O. O., & Sammut, C. (2010) · International Conference on Control, Automation, Robotics and Vision (ICARCV) · doi:10.1109/icarcv.2010.5707249 · Cited by 3
Space-variant kernel deconvolution for dual exposure problem
Tallón, M., Mateos, J., Babacan, S. D., et al. (2011) · Cited by 3
Task-based visual saliency for intelligent compression
Harding, P., & Roberston, N. (2009) · IEEE International Conference on Signal and Image Processing Applications (ICSIPA) · doi:10.1109/icsipa.2009.5478703 · Cited by 3
Continuous Background Update and Object Detection with Non-static Cameras
Zhao, Y., Casares, M., & Velipasalar, S. (2008) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2008.30 · Cited by 3
“As future work, we will incorporate Speeded Up Robust Features (SURF) [3] into our system to increase the processing speed.”
MethodLocal Part Model for Action Recognition in Realistic Videos
Shi, F. (2014) · uO Research (University of Ottawa) · doi:10.20381/ruor-3757 · Cited by 3
Using Computer Vision Technologies to Make the Virtual Visible
Büttner, S., Cai, T., Cramer, H. S. M., et al. (2011) · KTH Publication Database DiVA (KTH Royal Institute of Technology) · Cited by 3
Foreground Objects Recognition in Video Based on Bag-of-Words Model
Hu, M.-J., Li, C., Qu, Y., & Huang, J. (2009) · Chinese Conference on Pattern Recognition · doi:10.1109/ccpr.2009.5344073 · Cited by 3
Image prior combination in space-variant blur deconvolution for the dual exposure problem
Tallón, M., Mateos, J., Molina, R., et al. (2011) · Cited by 3
Art Forgery Detection via Craquelure Pattern Matching
Taylor, J. R. B., Baradarani, A., & Maev, R. G. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-20125-2_15 · Cited by 3
Fast Convex Layers Algorithm for Near-Duplicate Image Detection
Šinjur, S., Zazula, D., & Žalik, B. (2012) · Informatica · doi:10.15388/informatica.2012.379 · Cited by 3
Discrete Point Based Signatures and Applications to Document Matching
Spasojevic, N., Poncin, G., & Bloomberg, D. S. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-24085-0_16 · Cited by 3
“Many techniques like SURF [2] and SIFT [3], [4], FIT [5] perform well at point matching across images, and image lookup from a database.”
MethodNguyen, T. Q., Mann, G. K. I., & Gosine, R. G. (2014) · Canadian Conference on Computer and Robot Vision · doi:10.1109/crv.2014.50 · Cited by 3
Fast face detection and localization from multi-views using statistical approach
Anvar, S. M. H., Yau, W., & Teoh, E. K. (2011) · International Conference on Information, Communications & Signal Processing (ICICS) · doi:10.1109/icics.2011.6173610 · Cited by 3
Real-Time Cascade Template Matching for Object Instance Detection
Xie, C., Li, J., Wang, T., et al. (2012) · Springer eBooks · doi:10.1007/978-1-4614-3501-3_36 · Cited by 3
Design and Implementation of Automatic Aerial Mapping System Using Unmanned Aerial Vehicle Imagery
Javadi, M. S., Kadim, Z., & Woon, H. H. (2014) · Cited by 3
A region-dependent image matching method for image and video annotation
Abdollahian, G., Birinci, M., Díaz-de-María, F., et al. (2011) · International Workshop on Content-Based Multimedia Indexing (CBMI) · doi:10.1109/cbmi.2011.5972532 · Cited by 3
A New SIFT-Based Image Descriptor Applicable for Content Based Image Retrieval
Giveki, D., Soltanshahi, M. A., Shiri, F., et al. (2015) · Cited by 3
Social game epitome versus automatic visual analysis
Vajda, P., Ivanov, I., Lutz Goldmann, A., & Ebrahimi, T. (2011) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2011.6011927 · Cited by 3
“A faster version of SIFT descriptor with comparable accuracy, called Speeded Up Robust Features (SURF), is proposed in [9].”
MethodDifference of Circles Feature Detector
Hojaij, A., Fakih, A. H., Wong, A. M., & Zelek, J. (2012) · Canadian Conference on Computer and Robot Vision · doi:10.1109/crv.2012.16 · Cited by 4
Technologies for Granting Balance between Security and Privacy in Video-Surveillance
D'Arminio, P., Iglesias, R. B., Cichowski, J., et al. (2012) · European Intelligence and Security Informatics Conference · doi:10.1109/eisic.2012.38 · Cited by 3
“SURF) [15] as well as global features bas texture.”
BackgroundPl@ntNet's participation at LifeCLEF 2014 plant identifiation task
Goëau, H., Joly, A., Yahiaoui, I., et al. (2014) · Agritrop (Cirad) · Cited by 3
A web system for ontology-based multimedia annotation, browsing and search
Bertini, M., Becchi, G., Bimbo, A. D., et al. (2011) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2011.6012070 · Cited by 3
“[13] H. Bay, A. Ess, T. Tuytelaars, and L. Van Gool, “SURF: Speeded Up Robust Features,” Computer Vision and Image Understanding (CVIU), vol. 110, no. 3, pp. 346–359, 2008.”
MethodFast recursive ensemble convolution of Haar-like features
Węsierski, D., Mkhinini, M., Horain, P. J., et al. (2012) · 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.2012.6248115 · Cited by 3
Exploitation of 3D Information for Directing Visual Attention and Object Recognition
Akman, O., & Jonker, P. P. (2009) · Cited by 3
A new hierarchical approach in robust real-time image feature detection and matching
Langer, M., & Kuhnert, K. ‐. (2008) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2008.4761165 · Cited by 3
Numismatic Object Identification Using Fusion of Shape and Local Descriptors
Huber-Mörk, R., Zaharieva, M., & Czedik-Eysenberg, H. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-89646-3_36 · Cited by 3
“A descriptor is accepted only if the distance ratio to the second nearest neighbor is below a given threshold [26,25].”
BackgroundMethodTo Watch or Not to Watch: Video Summarization with Explicit Duplicate Elimination
Ouellet, J.-N., & Randrianarisoa, V. (2011) · Canadian Conference on Computer and Robot Vision · doi:10.1109/crv.2011.52 · Cited by 3
Descriptor evaluation and feature regression for multimodal image analysis
Yong, X., Yang, M. Y., Cao, Y., et al. (2015) · Machine Vision and Applications · doi:10.1007/s00138-015-0714-x · Cited by 3
“In this subsection, we will present four descriptors used in this paper, namely SIFT, SURF, LBP and HOG.”
MethodInfluentialAugmenting flower recognition by automatically expanding training data from web
Huang, C.-Y., Lin, Y.-L., & Hsu, W. H. (2015) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp.2015.7340809 · Cited by 4
NHK STRL at TRECVID 2009: Surveillance Event Detection and High-Level Feature Extraction.
Takahashi, M., Kawai, Y., Fujii, M., et al. (2009) · TRECVID · Cited by 3
Fast image matching algorithm based on affine invariants
Zhang, Y., Lü, K., & Gao, Y. (2014) · Journal of Central South University · doi:10.1007/s11771-014-2137-7 · Cited by 3
“[5], SURF [6], MSER [7] and RASIM [8]) focus on multiple interest points by computing some properties of the local image neighborhoods.”
BackgroundMOVING OBJECT TRACKING METHOD USING IMPROVED CAMSHIFT WITH SURF ALGORITHM
Joshi, S., Gujarathi, S., & Mirge, A. (2014) · Cited by 3
A stereo visual odometry based on SURF feature and three consecutive frames
Ren, Y., Xie, X., Hu, J., & Li, Z. (2015) · IEEE International Smart Cities Conference (ISC2) · doi:10.1109/isc2.2015.7366224 · Cited by 3
Rettkowski, J., Gburek, D., & Göhringer, D. (2015) · Conference on Design and Architectures for Signal and Image Processing · doi:10.1109/dasip.2015.7367240 · Cited by 3
“The image stitching function receives an array of input images and uses the SURF algorithm [18] to determine correlated points between the images.”
MethodAn evaluation of image registration methods for chest radiographs
Engin, M., Oğul, H., Ağıldere, A. M., & Sümer, E. (2015) · SAI Intelligent Systems Conference · doi:10.1109/intellisys.2015.7361237 · Cited by 3
TailoredBRIEF: Online per-feature descriptor customization
Richardson, A., & Olson, E. (2015) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2015.7353357 · Cited by 3
“This results in a very terse descriptor that can be matched much faster than alternatives like SURF [1], [12].”
MethodRecognizing Degraded Handwritten Characters
Diem, M., Sablatnig, R., Gau, M., et al. (2011) · Kölner Universitäts PublikationsServer (Universität zu Köln) · Cited by 3
A Graph Matching and Energy Minimization Based Algorithm for Lunar Surface Image Mosaic
Li, C., Liu, Z., Yang, X., Qiao, H., & Liu, C. (2015) · Communications in computer and information science · doi:10.1007/978-3-662-48558-3_5 · Cited by 3
Efficient Bone Detector and Geometric Descriptor for X-Ray Imaging
Romanowski, J., Korytkowski, M., & Scherer, R. (2015) · IEEE Symposium Series on Computational Intelligence (SSCI) · doi:10.1109/ssci.2015.191 · Cited by 3
“Describing bones by geometric shapes is much more useful than local interest point descriptors obtained by e.g. the SURF [2][7] or the SIFT [11] algorithms.”
MethodCapturing and tracking of building area based on structure saliency in airborne remote sensing video
Feng, J. Q., Bi, F., Chen, L., et al. (2015) · Science China Information Sciences · doi:10.1007/s11432-015-5291-0 · Cited by 3
“Comparison experiments with standard SIFT and SURF[6] have also been presented in Matlab 2013a to show the performance of our system.”
MethodAn Approach to Large Scale Interactive Retrieval of Cultural Heritage
Takami, M., Bell, P. R. F., & Ommer, B. (2014) · Eurographics · doi:10.2312/gch.20141307 · Cited by 4
“Typically, high-dimensional features like Scale-invariant feature transform (SIFT) [Low04] or related features [BTVG06,TL09] are used here, because of their robustness against small shifts, scale changes and affine transformations, which enables them to detect even objects taken from different…”
BackgroundIntegration of visual and depth information for vehicle detection
Makris, A., Perrollaz, M., Paromtchik, I. E., & Laugier, C. (2011) · HAL · Cited by 3
A Robust Method for Indoor Localization Using Wi-Fi and SURF Based Image Fingerprint Registration
Niu, J., Ramana, K. V., Wang, B., et al. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-07425-2_26 · Cited by 3
“– Computation of the Box Hessian operator [17] at several scales and sample rates using box-filters.”
MethodEfficient Panorama Mosaicing Based on Enhanced-FAST and Graph Cuts
Wang, X., Sun, J., & Peng, H. (2012) · Lecture notes in electrical engineering · doi:10.1007/978-3-642-25792-6_115 · Cited by 3
Lin, C.-T., Chen, L.-T., & Wang, Y.-F. (2015) · IEEE Vehicular Technology Conference · doi:10.1109/vtcfall.2015.7390937 · Cited by 3
“While SURF and USURF had smaller directional errors than SIFT, it might be because they had the lowest detection density (63% lower than ZBRIEF and SIFT) that allowed them to focus on fewer, highquality features (the public-domain program [7,20] did not allow the feature density to increase beyond what we used).”
BackgroundMethodFocal-Plane Scale Space Generation with a 6T Pixel Architecture
Oliveira, F. D. V. R., Gomes, J. G. R. C., Carmona‐Galán, R., et al. (2016) · Electronic Imaging · doi:10.2352/issn.2470-1173.2016.12.imse-270 · Cited by 3
Small Moving Object Tracking in Dynamic Video
Sun, W., Li, D., Jia, W., Li, P., Zhao, C., & Chen, X. (2015) · International Conference on Intelligent Information Hiding and Multimedia Signal Processing (IIH-MSP) · doi:10.1109/iih-msp.2015.25 · Cited by 3
Adaptive visual target detection and tracking using incremental appearance learning
Yazdian‐Dehkordi, M., & Azimifar, Z. (2015) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2015.7350958 · Cited by 3
Feature Point Tracking Algorithm Evaluation for Augmented Reality in Handheld Devices
Perera, A., Pemasiri, A., Wijayarathna, S., et al. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16631-5_21 · Cited by 3
“These trackers use different combinations of SIFT and SURF feature detector and descriptor.”
MethodInfluentialScene Classification by Feature Co-occurrence Matrix
Lang, H., Xi, Y., Hu, J., et al. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16628-5_36 · Cited by 3
“Among the vastly employed methods are the scale-invariant feature transform(SIFT) [4], speeded up robust feature(SURF) [5], histogram of oriented gradients(HOG) [6], gradient location and orientation histogram(GLOH) [7], region covariance matrix(RCM) [8], local binary patterns(LBP) [9] etc.”
MethodTextureless object detection using cumulative orientation feature
Konishi, Y., Ijiri, Y., Suwa, M., & Kawade, M. (2015) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2015.7351012 · Cited by 4
Evaluation of Interest Point Detectors in Presence of Noise
Ziomek, A., & Oszust, M. (2016) · International Journal of Intelligent Systems and Applications · doi:10.5815/ijisa.2016.03.03 · Cited by 3
Temporal and Spatial Evolution through Images
Branco, F., Correia, N. M. R., Rodrigues, A., et al. (2015) · IEEE International Symposium on Multimedia (ISM) · doi:10.1109/ism.2015.105 · Cited by 3
“The algorithms considered to be included in Evolapse were: SIFT (Scale-Invariant Feature Transform) [10], SURF (Speeded Up Robust Features) [4], FAST (FAST Corner Detector) [16] and ORB (ORiented Brief) [13].”
MethodMedical image registration using extremal region based interest points
Kashyap, M., & Bhattacharya, M. (2015) · IEEE International Conference on Computer Graphics, Vision and Information Security (CGVIS) · doi:10.1109/cgvis.2015.7449929 · Cited by 3
“Fig 3c and 3d are the results of registering 3a and 3b by using MSER based IPs and the proposed ER based IPs respectively (using Speeded up Robust Features [19-20]).”
MethodUnsupervised Region of Intrest Detection Using Fast and Surf
Olaode, A. A., Naghdy, G. A., & Todd, C. (2015) · doi:10.5121/csit.2015.50406 · Cited by 3
Development of Real-Time Markerless Augmented Reality System Using Multi-thread Design Patterns
Jin, D.-X., Um, K., & Cho, K. (2011) · Communications in computer and information science · doi:10.1007/978-3-642-27204-2_18 · Cited by 3
“In 2006, Bay [3] proposed a SURF algorithm that improved the speed of the SIFT algorithm.”
MethodImproved global context descriptor for describing interest regions
Liu, J., & Zeng, G. (2012) · Journal of Shanghai Jiaotong University (Science) · doi:10.1007/s12204-012-1244-6 · Cited by 3
“To compute the GC descriptor for SIFT, Carmichael et al. [11] proposed a similar GC descriptor for speeded up robust feature (SURF) [12] and maximally stable extremal region (MSER) [13] .”
MethodPrecise 3D recording for finite element analysis
Kalisperakis, I., Stentoumis, C., Γραμματικόπουλος, Λ., et al. (2015) · 2015 Digital Heritage · doi:10.1109/digitalheritage.2015.7419467 · Cited by 3
Texture classification using an invariant texture representation and a tree matching kernel
Soottitantawat, S. (2010) · doi:10.14457/nida.the.2010.54 · Cited by 3
A Short Overview of Feature Extractors for Knuckle Biometrics
Choraś, M. (2013) · Advances in intelligent systems and computing · doi:10.1007/978-3-319-00969-8_51 · Cited by 3
Kri, J., & Dobri, S. (2012) · Cited by 4
Descriptor Trends in Texture Classification for Material Recognition
Ayad, H., Abdulameer, M. H., George, L. E., & Hassan, N. F. (2015) · Research Journal of Applied Sciences Engineering and Technology · doi:10.19026/rjaset.10.1878 · Cited by 3
Matching of Complex Scenes Based on Constrained Clustering.
Loui, A. C. P., & Das, M. (2008) · National Conference on Artificial Intelligence · Cited by 3
PERFORMANCE EVALUATION OF FEATURE DETECTION FOR LOCAL OPTICAL FLOW TRACKING
Senst, T., Unger, B., Keller, I., & Sikora, T. (2012) · International Conference on Pattern Recognition Applications and Methods · doi:10.5220/0003731103030309 · Cited by 3
Fusing Time-of-Flight Cameras and Inertial Measurement Units for Ego-Motion Estimation
Droeschel, D., May, S., Holz, D., et al. (2011) · Automatika · doi:10.1080/00051144.2011.11828419 · Cited by 3
“Various refinements of the basic SIFT algorithm have been proposed, e.g. PCA-SIFT [18], GLOH [19], and SURF [20].”
MethodA Robust Approach for Action Recognition Based on Spatio-Temporal Features in RGB-D Sequences
Quoc, L., Hoai, V., Thai, T., & Minh, P. (2016) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2016.070526 · Cited by 3
The New Approach to Camera Calibration – GCPs or TLS Data?
Markiewicz, J. S., Podlasiak, P., Kowalczyk, M., & Zawieska, D. (2016) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xli-b3-75-2016 · Cited by 5
“…for corresponding natural points in the images and orthoimages (intensity and shaded DSM), the algorithms based on corner detection were used: BRISK (Leutenegeer et al., 2011), FAST (Rostem, 2006) and BLOB algorithms: MSER (Matas et al., 2002), SIFT (Lowe, 2004), SURF (Bay et al., 2006).”
MethodSURF-based mammalian species identification system
Otero, B., Rodríguez, E., & Ventura, J. (2016) · Multimedia Tools and Applications · doi:10.1007/s11042-016-3602-0 · Cited by 3
Pose and illumination invariant face recognition using binocular stereo 3D reconstruction
Nigam, A., Chhalotre, G., & Gupta, P. (2015) · National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG) · doi:10.1109/ncvpripg.2015.7489941 · Cited by 4
“The SURF score level fusion of three scores obtained from AHE, HGORP and SF enhancements is proposed.”
MethodInfluentialDIRECT IMAGE-TO-GEOMETRY REGISTRATION USING MOBILE SENSOR DATA
Kehl, C., Buckley, S. J., Gawthorpe, R. L., et al. (2016) · ISPRS annals of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-annals-iii-2-121-2016 · Cited by 3
“After obtaining the photo and the corresponding synthetic image, a feature detector (e.g. SIFT (Lowe, 2004), Speeded-Up Robust Features (SURF) (Bay et al., 2006), Maximally Stable Colour Regions (MSCR) (Forssén, 2007)) extracts 2D keypoints and descriptors.”
MethodIP Camera Based Video Surveillance Using Object’s Boundary Specification
Chaudhry, N., & Suleman, K. M. U. (2016) · International Journal of Information Technology and Computer Science · doi:10.5815/ijitcs.2016.08.02 · Cited by 3
“[11] We take current frame as query frame and image of registered object and then apply SURF feature matching algorithm on these two images.”
MethodVisual Search for Objects in a Complex Visual Context: What We Wish to See
Boujut, H., Bugeau, A., & Benois‐Pineau, J. (2016) · CRC Press eBooks · doi:10.1201/b17080-2 · Cited by 3
Etemad, E., Hu, G., & Gao, Q. (2016) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2016.7532408 · Cited by 3
“There are various approaches for local feature description, some of which detect keypoints such as FAST [4], and Harris corner detectors [5], some provide local descriptors, such as BRIEF[6] and FREAK [7], and some perform both detection and description tasks such as SIFT [8], SURF [9], ORB [10] and BRISK [11].”
MethodA System to Help Amateurs Take Pictures of Delicious Looking Food
Kakimori, T., Okabe, M., Yanai, ., & Onai, R. (2016) · IEEE International Conference on Multimedia Big Data · doi:10.1109/bigmm.2016.47 · Cited by 3
“For example, histograms of colors and image gradients remain popular and powerful features, e.g., SIFT [3], SURF [4], and ORB features [5].”
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.