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 12,301–12,400 of 14,752
A Method to Standing Tree Image Mosaic Based on SURF in Pocket PC
Han, D. (2012) · International Conference on Computer Science and Electronics Engineering (ICCSEE) · doi:10.1109/iccsee.2012.106
Object edge contour localisation based on HexBinary feature matching
Liu, Y., Aragón-Camarasa, G., & Siebert, J. P. (2014) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio.2014.7090314
Embedding class information into local invariant features by low-dimensional retinotopic mapping
Raytchev, B., Kikutsugi, Y., Tamaki, T., et al. (2012) · Machine Vision and Applications · doi:10.1007/s00138-012-0415-7
A pattern matching scheme for accurate medicine localization in automatic pharmacy inventory
Han, D. S., & Menga, G. (2012) · IEEE International Conference on Computer Science and Automation Engineering · doi:10.1109/csae.2012.6272937
“Most training involved methods extract image-content based local features, such as SIFT [5], SURF [6], HOG [7] or texture information [8] to characterize objects and then train classifiers like Bayesian [9], SVM [10] and HMM [11], to determine pixel categories.”
MethodMultilayer Semantic Analysis in Image Databases
Sayad, I. A. E., Martinet, J., Zhang, Z. M., et al. (2014) · Annals of information systems · doi:10.1007/978-3-319-07812-0_19
Hybrid multiple-object tracker incorporating Particle Swarm Optimization and Particle Filter
Hsu, C.-C. J., Chu, Y.-C., & Lu, M.-C. (2013) · International Conference on System Science and Engineering · doi:10.1109/icsse.2013.6614657
Letouzey, A., Petit, B., & Boyer, E. (2012) · Traitement du signal · doi:10.3166/ts.29.255-281
“tantes pour calculer des correspondances 2D entre une paire d’images, par exemple, SIFT (Lowe, 2004), SURF (Bay et al., 2006) ou Harris (Harris, Stephens, 1988).”
Second Screen Solution Based on Cloud Centric Media Platform
Xie, T., & Jin, Y. (2014) · Applied Mechanics and Materials · doi:10.4028/www.scientific.net/amm.513-517.1835
“After the system receives the image taken by the mobile device camera from users, we use Speeded-Up Robust Features (SURF) [1] to extract the visual features on the two images.”
MethodPeople reidentification in a distributed camera network
de Oliveira, Í. O., & Pio, J. L. D. S. (2009) · doi:10.1145/1858477.1858529
“The SURF (BAY et al., 2006) descriptor acknowledges the high matching performance of SIFT (LOWE, 2003) and attempts to replicate this performance while improving speed.”
BackgroundInfluentialImage Retrieval Based on Two-Tuples Histogram
Liu, G. (2010) · Chinese Conference on Pattern Recognition · doi:10.1109/ccpr.2010.5659152
“的影响。经典的形状特征描述方法主要包括傅立叶变 换,曲率尺度空间和Zernike矩([1])等形状特征描述子。在 提取形状特征之前,都需要进行高效的图像分割。从自 然图像中尤其在包含有原始图像的大型图像库中,提取 真正的形状信息仍然是一个公开的难题。全局特征描述 子在描述图像内容方面具有一定的优势,但是它们不太 适合于描述图像中的实际对象。目前比较流行利用局部 特征来描述图像内容,经典的局部特征描述子主要有 SIFT([12]),SURF([14]),GLOH([13])以及PCA-SIFT([15])等描述子。”
Improving viewpoint invariance of image feature extraction methods using intensity and range images
Kovács, V., & Tevesz, G. (2012) · International Carpathian Control Conference · doi:10.1109/carpathiancc.2012.6228670
“The most common feature extractors such as SIFT [3] (Scale Invariant Feature Transform) and SURF [7] (Speeded Up Robust Feature) provide robust feature detection independently of changes in translation, rotation and also reject small changes in illumination and viewpoint variance.”
MethodResultMarkerless monocular tracking system for guided external eye surgery
Monserrat, C., Rupérez, M. J., Alcañíz, M., & Mataix, J. (2014) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2014.08.001 · Cited by 2
“The most widely used current algorithms in computing invariant points and their correspondence between pairs of images are: the Scalable Invariant Feature Transforms (SIFT) [28] and the Speed Up Robust Features (SURF) [29].”
MethodImage-Based Localization in the Wireless Sensor Network
Baek, S. M., & Kim, D. (2013) · IFAC Proceedings Volumes · doi:10.3182/20131111-3-kr-2043.00021
“We use Speed Up Robust Features(SURF) algorithm to solve that correspondence problem (Bay et al., 2006).”
MethodGPS-denied hover stabilization of a quad-rotor via visual servoing
Buck, L. H., Manegold, D., & Ray, L. E. (2014) · IEEE International Conference on Technologies for Practical Robot Applications · doi:10.1109/tepra.2014.6869147
Image processing system dedicated to a visual intra‐cortical stimulator
Ghannoum, A., Ghafar‐Zadeh, E., & Sawan, M. (2014) · IET Image Processing · doi:10.1049/iet-ipr.2013.0838
Towards automatic identification of mismatched image pairs through loop constraints
Elibol, A., Kim, J., Gracias, N., et al. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2040778
“To register images, these methods follow a pipeline that is composed of (1) feature detection, (2) feature descr iption and (3) descriptor matching.”
MethodSurveillance Issues in a Smart Home Environment
Tsitsoulis, A., Patrick, R., & Bourbakis, N. (2012) · Proceedings - International Conference on Tools with Artificial Intelligence, TAI · doi:10.1109/ictai.2012.182
Field of view alignment with multiple RGB-D cameras
Hu, W.-C., & Lin, C. H. (2013) · IEEE International Symposium on Consumer Electronics (ISCE) · doi:10.1109/isce.2013.6570166
A two-stage matching algorithm on heterogeneous platforms for detecting Advertisement Logo in videos
Cao, X., Xu, C., & Xiu-hua, D. (2014) · Chinese Control Conference (CCC) · doi:10.1109/chicc.2014.6895739
“3 SURF [5] DOT 480P 720P 4 DOT SURF 480P 720P SURF”
InfluentialToward a practical visual object recognition system
Nguyen, M., & Tran, M. (2013) · Symposium on Information and Communication Technology (SoICT) · doi:10.1145/2542050.2542077
Classification and indexing in highway traffic videos
Kamath, S. V., Darbari, M., & Shettar, R. (2013) · International Conference on Contemporary Computing · doi:10.1109/ic3.2013.6612220
Gradual Block-based Efficient Lossy Location Coding for Image Retrieval
Choi, G., Jung, H.-I., & Kim, H.-K. (2013) · Journal of Broadcast Engineering · doi:10.5909/jbe.2013.18.2.319
“이에 따라 최근 급속도로 발 전하는 모바일 환경에 대비하여 빠른 전송과 정확한 매칭 을 위한 강인한 디스크립터 기반의 SIFT(scale invariant feature transform)([3])와 SURF(Speeded up robust features) [4] 같은 알고리즘들의 특징 점 정보의 압축 알고리즘들이 많 이 제안되었다(그림 1).”
Topological localization based on key-frames selection and vocabulary tree for mobile robots
Xiong, D., Lu, H., Zeng, Z., et al. (2013) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio.2013.6739848
“The Speeded Up Robust Features (SURF) is an algorithm that detects key points based on the Hessian matrix and integral image[8].”
A Distributed Augmented Reality System Using 3D Fiducial Objects
Montgomery, B., Wu, Q., & Gu, Y. (2008) · International Conference on Cyberworlds (CW) · doi:10.1109/cw.2008.31
Intensity comparison based compact descriptor for mobile visual search
Na, S.-I., Lee, K., Lee, S., et al. (2013) · doi:10.1109/fcv.2013.6485469
Scale- and rotation- invariant feature detectors on Cellular Processor Arrays
Fernandez, N. A., Brea, V. M., Suarez, M., et al. (2012) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas.2012.6271853
A fast image rectification method for video captured with UAV hovering over traffic intersections
Xu, T., Wang, J., & Li, X. (2015) · Annals of GIS · doi:10.1080/19475683.2014.992371
A unified formalism for video descriptors
Kihl, O., Picard, D., & Gosselin, P.-H. (2013) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2013.6738498
Ishii, A., Sakai, A., Mitsuhashi, M., et al. (2010) · IFAC Proceedings Volumes · doi:10.3182/20100913-3-us-2015.00064
A real-time orientation feature descriptor for portable devices
Yang, P., Jiang, Z., Feng, H., et al. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2051141
“SURF achieved the best performance in this experiment.”
BackgroundInfluentialGrid features based visual tracking
Zhou, Y., Snoussi, H., & Zheng, S. (2011) · IEEE International Conference on Computer Science and Automation Engineering · doi:10.1109/csae.2011.5952844
Vision-Based Path Learning for Home Robots
Ueno, A., Kajihara, N., Fujii, N., et al. (2014) · International Conference on Intelligent Information Hiding and Multimedia Signal Processing (IIH-MSP) · doi:10.1109/iih-msp.2014.109
“Then, the method uses a BoVW technique [3] to create the input space and an initial set of states: All features in the first Thimage images in Random-Walk Phase are detected and described by the SURF descriptor [4], visual words are defined by clustering all the features, the space of the normalized histograms of them…”
MethodTarget location in aerial image by fast NCC
Dong, J., Yang, X., & Hou, W. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2054151
“where i t % and i b% are given in formula (2), where t and b are the mean value of vectors t and b .”
BackgroundMultimedia Analysis of Video Sources
Irujo, J. A., Cuadros, M., Aginako, N., et al. (2014) · International Conference on Signal Processing and Multimedia Applications · doi:10.5220/0005126903460352
Exact and easy guidance with visual navigation situation for mobile user
Huang, Z., Fang, H., Tuo, X.-Y., et al. (2013) · International Conference on Internet Multimedia Computing and Service · doi:10.1145/2499788.2499840
AnySURF: Flexible Local Features Computation
Sadeh-Or, E., & Kaminka, G. A. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-27216-5_17
“Unfortunately, existing algorithms for local feature matching [2,11,12] are designed under the assumption that they will run to completion and only then”
BackgroundMethodInfluentialScale-Adaptive Feature Points Extraction and Dual-Scale Based High-Precision Image Registration
Gao, Z., Ming, D., Xiao, L., et al. (2012) · IEEE International Conference on Computer and Information Technology (CIT) · doi:10.1109/cit.2012.33
Case deletion for fundamental matrix computation
McCane, B. (2014) · International Conference on Image and Vision Computing New Zealand (IVCNZ) · doi:10.1145/2683405.2683416
“Despite the amount of research on fast detectors/descriptors, it appears that most researchers tend to still use SIFT [11] or its faster variant SURF [1].”
BackgroundImage-based motion estimation of underwater tow body
Wang, C., Chen, H.-H., & Li, K.-H. (2014) · doi:10.1109/oceans-taipei.2014.6964571
Search algorithm complexity modeling with application to image alignment and matching
DelMarco, S. P. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2049605
A Robust Affine Invariant Feature Matching Approach
Gao, C., Song, Y., & Jia, P. (2011) · doi:10.1109/icmla.2011.21
“SURF (Speeded Up Robust Features)[7] is another famous descriptor proposed by Herbert Bay et al.”
BackgroundInfluentialString Features: Geodesic Sweeping Detection and Quasi-invariant Time-Series Description
Guerra‐Filho, G. (2012) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2012.72
“, SURF feature detector) [1], and difference of Gaussians (i.”
BackgroundLunar orbiter automotive navigation based on camera and laser altimeter
Xia, J., Tang, G., Bu, Y., et al. (2014) · doi:10.1109/cgncc.2014.7007532
Real Time Method to Detect the Waving Drive from a Single Camera
Teshima, T., Saitô, H., Ozawa, S., et al. (2009) · IEEJ Transactions on Electronics Information and Systems · doi:10.1541/ieejeiss.129.1714
A Rapid and Efficient Feature Point Detection and Matching Algorithm
Ding, X. L., Zhao, Q., Li, Y. B., & Ma, X. (2014) · Applied Mechanics and Materials · doi:10.4028/www.scientific.net/amm.607.641
Objects Recognition and Intelligent Walking for Quadruped Robots based on Genetic Programming
Kim, Y., Hyun, S., Jang, J.-Y., et al. (2010) · Journal of Korean institute of intelligent systems · doi:10.5391/jkiis.2010.20.5.603
“특징점을 통해 영상을 인식하는 방법 은 크게 특징점 추출과 표현자(descriptor) 생성 과정이 필 요하며, 대표적인 방법으로는 스케일 불변 특징 변환 (SIFT:Scale Invariant Feature Transform)[5] 알고리즘과 강인한 고속 특징 추출(SURF:Speeded Up Robust Features)[2] 알고리즘이 있다.”
Beyond semantic attributes: Auxiliary feature discovery for image classification
Niu, B., Cheng, J., Liu, Y., et al. (2014) · Neurocomputing · doi:10.1016/j.neucom.2014.03.063
OpenSURF Algorithm Improvement for Augmented Reality Based on the Natural Features
Yang, J., Fang-wen, Z., Yuan, Z., et al. (2012) · International Conference on Intelligent Human-Machine Systems and Cybernetics · doi:10.1109/ihmsc.2012.21
“Based on the SURF algorithm, open source project OpenSURF [7] [8] is developed.”
BackgroundMethodInfluentialLocal binary feature based on census transform for mobile robot
Touma, Y., Hagiwara, H., Asami, K., et al. (2014) · Joint International Conference on Soft Computing and Intelligent Systems and International Symposium on Advanced Intelligent Systems (SCIS&ISIS) · doi:10.1109/scis-isis.2014.7044645
A Novel Progressive Transmission in Mobile Visual Search
Zhang, Q., Li, Z., Du, Y., et al. (2014) · IEEE International Conference on Dependable, Autonomic and Secure Computing · doi:10.1109/dasc.2014.53
Hybrid map-building and localization for unstructured and moderately dynamic environments
Rady, S., & Badreddin, E. (2012) · IEEE International Conference on Intelligent Systems · doi:10.1109/is.2012.6335165
“Local point feature extraction [4,5] is a suitable technique that suits modeling unstructured environments.”
BackgroundMethodControl Point Extraction in the Remote Sensing Image via Adaptive Filter
Geng, L., Xia, D. S., Sun, Q. S., et al. (2013) · Applied Mechanics and Materials · doi:10.4028/www.scientific.net/amm.411-414.1267
“Recently, the researches focus on Harris operator [4], Forstner operator [5], scale invariant features transform algorithm [6] and speeded up robust features (Surf) algorithm [7, 8].”
BackgroundFeature detection and SLAM on embedded processors for micro-robot navigation
Robinette, P., & Collins, T. (2013) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2015651
“Feature Detection and Description Using SURF Speeded Up Robust Features (SURF) uses difference of Gaussians to determine which pixels in an image belong to a feature.(3,4,5) This is essentially a second order derivative comparison of pixels around the candidate interest point to determine if there is a sufficient inte…”
MethodLocal Covariant Region Detection Based on Image Contour Corner
Xu, L., Yang, M., Wang, H., et al. (2012) · International Symposium on Information Science and Engineering · doi:10.1109/isise.2012.58
An Interest Point Detector Based on Polynomial Local Orientation Tensor
Lin, R., Wang, W., Du, Z., et al. (2012) · Acta Automatica Sinica · doi:10.1016/s1874-1029(11)60290-5
An object classification framework based on unmeasurable area patterns found in 3D range images
Matsumoto, K., & Yamazaki, K. (2014) · 2022 IEEE 18th International Conference on Automation Science and Engineering (CASE) · doi:10.1109/coase.2014.6899333
“SIFT [8] and SURF [9], and others [10] are useful for them.”
MethodFacilitating direct and ubiquitous mobile computer vision
Wirtz, H., Rüth, J., & Wehrle, K. (2014) · International Conference on Mobile and Ubiquitous Multimedia (MUM) · doi:10.1145/2677972.2677974
Dynamic learning, retrieval, and tracking to augment hundreds of photographs
Pilet, J., & Saitô, H. (2013) · Virtual Reality · doi:10.1007/s10055-013-0228-7
Registration of cardiac magnetic resonance images using SURF points and matching
Moreno, B., Atehortúa, A., Corredor, G., et al. (2015) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2073527
Pooling Search: Serum Samples Test Simulated Video Fingerprint Search
Yao, J., Yu, H., & Hu, R. (2012) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2012.151
Enhanced point descriptors for dense stereo matching
Lang, H., Wang, Y., Qi, X., et al. (2010) · International Conference on Image Analysis and Signal Processing · doi:10.1109/iasp.2010.5476124
Voting-Based Matching for Near-Planar Objects with Drastic Illumination Variation
Xie, H., & Yan, J. (2010) · doi:10.1109/icmult.2010.5631353
Susaki, J., Mishra, B., & Ota, Y. (2013) · Joint Urban Remote Sensing Event (JURSE) · doi:10.1109/jurse.2013.6550652
LINE-BASED MULTI-IMAGE MATCHING FOR FAÇADE RECONSTRUCTION
Teo, T., & Kao, C.-H. (2012) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprsarchives-xxxix-b3-63-2012
“In the orientation determination, Speeded Up Robust Features (SURF) is applied to extract tie points automatically.”
BackgroundInfluentialA Novel Method to Automatically Generate Depth Maps Based on Trace Points
Zhao, T. Q., Sang, X. Z., Liu, Y. D., et al. (2014) · Applied Mechanics and Materials · doi:10.4028/www.scientific.net/amm.513-517.3890
“Some filter operators such as Harris [4]and SURF[5] are typically used because of their stability after linear deformation of image.”
Augmented Reality based Dynamic State Transition Algorithm using the 3-Axis Accelerometer Sensor
Jang, Y., & Park, S.-J. (2010) · The Journal of the Korea Contents Association · doi:10.5392/jkca.10.10.086
Distance Learning Using ExperInn E-Learning System Through Web
Shetty, C. G. (2012) · International Journal of Web-Based Learning and Teaching Technologies · doi:10.4018/jwltt.2012040104
Map-assisted visual localization using line features in urban area
Li, H., Wang, H., Lu, X., & Liu, J. (2012) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc.2012.6243064
“Point features, such as scale-invariant feature translation (SIFT)[4], and speed up robust feature (SURF) points [5] are the most popular and reliable ones.”
BackgroundA Method to Compare Images for Managing Tools to Repair Ships
Park, S. H., & Kim, J.-D. (2014) · The Journal of the Korean Institute of Information and Communication Engineering · doi:10.6109/jkiice.2014.18.10.2489
Image modeling system development for robust descriptor of environmental change
Choi, G. M., Junz, H. I., Kim, R. K., et al. (2013) · doi:10.1109/fcv.2013.6485461
Object detection using signature library
Priyanka, M., Faheema, A. G. J., & Rakshit, S. (2012) · International Conference on Computing, Communication and Networking Technologies (ICCCNT) · doi:10.1109/icccnt.2012.6396009
A Novel Image Signature based on Local Representative Pattern Mining
Wang, G., Gao, K., & Li, J. (2014) · doi:10.1145/2632856.2632924
“Currently available methods for copy detection in literature, such as SIFT [6] and SURF [2], show great performance under various image transformations, however, their high complexities in feature extracting and matching limit their use in large volume dataset of copy detection, so it is essential to propose a new met…”
Target extraction from SAR image under the guide of GIS
Liu, M., Guo, L., Lu, X., et al. (2014) · doi:10.1109/icosp.2014.7015135
Zhao, D., Wang, J. J., Wan, J., et al. (2013) · Applied Mechanics and Materials · doi:10.4028/www.scientific.net/amm.380-384.1235
Hierarchical Encoding of Binary Descriptors for Image Matching
Mao, Z., Tong, L., Xie, H., et al. (2015) · ACM International Conference on Multimedia Retrieval (ICMR) · doi:10.1145/2671188.2749304
“However, these new descriptors tend to be less robust and discriminative than slower approaches such as SIFT [13] and SURF [3].”
BackgroundMethodDirectional SIFT -- An Improved Method Using Elliptical Gaussian Pyramid
Xu, X., & Yang, J. J. (2010) · Chinese Conference on Pattern Recognition · doi:10.1109/ccpr.2010.5659135 · Cited by 1
Match elimination using cycle basis in underwater optical mapping
Elibol, A., Kim, J., Yu, S., Gracias, N., & García, R. (2013) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceans-bergen.2013.6608034
A Two-Layer Descriptor and Two-Step Matching for Stereoscopic Images
Lin, C.-L., Wang, L., Li, D., et al. (2013) · International Conference on Instrumentation and Measurement, Computer, Communication and Control (IMCCC) · doi:10.1109/imccc.2013.310
“The most famous feature detector and descriptor are the Scale Invariant Feature Transform (SIFT) keypoints proposed by Lowe [4] and the Speed Up Robust Feature (SURF) by Bay [5].”
MethodZhang, Z., Ai, M., Zhang, Z., et al. (2010) · International Conference on Geoinformatics · doi:10.1109/geoinformatics.2010.5568083
“[5] Bay, H., T. Tuytelaars, L.V. Gool, SURF: Speeded Up Robust Features, in Proceedings of the European Conference on Computer Vision.”
MethodInfluentialGatter, A., Doehler, H.-U., Lueken, T., et al. (2014) · CEAS Aeronautical Journal · doi:10.1007/s13272-014-0135-y
A Method of Image Registration Based On Best Similarity of Local Geometric Figure
Jiang, Z., Zhou, L., & Zhang, L. (2014) · Journal of Software · doi:10.4304/jsw.9.4.964-969
Real-time Video Mosaicing using Non-rigid Registration
de Souza, R. H. C., Okutomi, M., & Torii, A. (2012) · IPSJ Transactions on Computer Vision and Applications · doi:10.2197/ipsjtcva.4.170
A Robust Feature Matching Approach for Photography Originality Test
Gao, C., Song, Y., & Jia, P. (2011) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-011-0321-z
“SIFT spends 1190 ms on every picture and PCA-sift also needs 237 ms. SURF is always deemed as a real-time descriptor.”
Design of a Passive System for Human Body Reconstruction in the Fashion Industry
Forlani, G., Nardinocchi, C., & Roncella, R. (2010) · International Conference on 3D Body Scanning Technologies (3DBST) · doi:10.15221/10.296
“However in this case such points are not measured on the Region of Interest (ROI); therefore a procedure often used on structure from motion techniques is applied: image features are extracted and matched using the SURF operator [1] and the known epipolar geometry.”
MethodData independent visual vocabulary
Faheema, A. G. J., Lakshmi, A., Priyanka, M., et al. (2012) · International Conference on Signal Processing and Communications (SPCOM) · doi:10.1109/spcom.2012.6290027
Fuangkaew, S., & Patanukhom, K. (2015) · doi:10.4108/icst.iniscom.2015.258352
“After the boundaries of the object are found in the scene image, a local feature matching such as Scale Invariant Feature Transform (SIFT) [9], Speed-Up Robust Feature (SURF) [10] is applied to extract the keypoints inside the object boundaries.”
MethodRobust localization using RGB-D images
Oh, Y., & Oh, S. (2014) · International Conference on Control, Automation and Systems (ICCAS) · doi:10.1109/iccas.2014.6987940
Efficient Point Feature Tracking based on Self-aware Distance Transform
Park, M.-G., & Yoon, K. (2012) · British Machine Vision Conference (BMVC) · doi:10.5244/c.26.32
A Dynamic Image Matching Model and Architecture for Smart Devices
Shahid, T., Ahmed, A. U., Craney, C., et al. (2011) · doi:10.1109/ithings/cpscom.2011.113
“The dimension includes some modifications and extensions of the existing matching methods [7], and algorithms [4].”
BackgroundMethodInfluentialLearned and designed features for sparse coding in image classification
Doan, D. A., Tran, N.-T., Vo, P. D., & Le, B. H. (2013) · RIVF International Conference on Computing and Communication Technologies (RIVF) · doi:10.1109/rivf.2013.6719900
“This is because there are some designed features (for example: SIFT [3], SURF [4], DAISY [5], etc), which can represent orientation, hence these features are able to support DN descriptors in image representation.”
MethodScene recognition with bag of visual nouns and prepositions
Stalbaum, J., Chae, H.-W., & Song, J.-B. (2015) · Intelligent Service Robotics · doi:10.1007/s11370-015-0167-0
A target recognition matching method with high reliability
Chen, S., Liu, Z., Xia, M., et al. (2011) · doi:10.1109/emeit.2011.6024094
A Multi-Stage Framework for Classification of Unconstrained Image Data from Mobile Phones
Mujumdar, S., Porat, D., Rajamani, N., & Subramaniam, L. V. (2014) · International Journal of Multimedia Data Engineering and Management · doi:10.4018/ijmdem.2014100102
A Multi-Level Scene Matching Algorithm Based on Adaptive Polar Transform
Lei, B., Fu, Y., Cheng, Y., et al. (2010) · doi:10.1109/cise.2010.5677222
“The so called SIFT [1] feature and the SURF [2] feature are invariant to illumination, scale, and rotation, and matching methods based on these features are robust to illumination, scale and rotation.”
MethodWang, Y., Duan, L., Lin, J., et al. (2014) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2014.7025808
AR Based on ORB Feature and KLT Tracking
Ren, J., Yu, Y., Zhou, Y., et al. (2013) · Applied Mechanics and Materials · doi:10.4028/www.scientific.net/amm.339.333
“One is based on the absolute value such as SIFT [9] and SURF [2] which are very accurate but need a lot of computing cost.”
MethodA Filtration Strategy Based on FD-CRF for Image Matching
Huang, S., & Wang, W. (2014) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2014.566
“In recent years, the local descriptors such as Scale Invariant Feature Transform (SIFT) [9], Speeded Up Robust Features (SURF) [10], and Gradient Location Orientation Histogram (GLOH) [11] are widely used due to the excellent performance shown in the experiments.”
MethodAirborne traffic monitoring supported by fast calculated digital surface models
Tuermer, S., Kurz, F., Reinartz, P. H., et al. (2012) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss.2012.6352593
“Therefore two consecutively imaged road segments are matched with the SURF operator [15].”
MethodDependence of gradient moment based descriptors on affine distortions of the differentiating kernel
Prohászka, Z. (2011) · IEEE International Symposium on Intelligent Systems and Informatics (SISY) · doi:10.1109/sisy.2011.6034326
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.