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 3,101–3,200 of 14,752
Accurate shape-based 6-DoF pose estimation of single-colored objects
Azad, P., Asfour, T., & Dillmann, R. (2009) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2009.5354606 · Cited by 22
An interactive augmented reality coloring book
Clark, A., Duenser, A., & Grasset, R. (2011) · SIGGRAPH Asia Emerging Technologies · doi:10.1145/2073370.2073394 · Cited by 24
Adapting New Categories for Food Recognition with Deep Representation
Ao, S., & Ling, C. X. (2015) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw.2015.203 · Cited by 21
Thongkor, K., Amornraksa, T., & Delp, E. J. (2018) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2018.03.005 · Cited by 22
Robust Object-Based Watermarking Using Feature Matching
PHAM, V.-Q., Miyaki, T., YAMASAKI, T., et al. (2008) · IEICE Transactions on Information and Systems · doi:10.1093/ietisy/e91-d.7.2027 · Cited by 22
Geometric VLAD for Large Scale Image SearchPreprint
Wang, Z., Di, W., Bhardwaj, A., Jagadeesh, V., & Piramuthu, R. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1403.3829 · Cited by 23
Transitive Invariance for Self-supervised Visual Representation LearningPreprint
Wang, X., He, K., & Gupta, A. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1708.02901 · Cited by 22
LPD-Net: 3D Point Cloud Learning for Large-Scale Place Recognition and Environment AnalysisPreprint
Liu, Z., Zhou, S., Suo, C., et al. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.07050 · Cited by 22
“Many successful image-based approaches were proposed in the literature, due to the feasibility of extracting visual feature descriptors (e.g. SURF [6] or ORB [16]).”
BackgroundTake a Look Around: Using Street View and Satellite Images to Estimate House PricesPreprint
Law, S., Paige, B., & Russell, C. (2018) · RePEc: Research Papers in Economics · doi:10.48550/arxiv.1807.07155 · Cited by 32
“The result found objects identified using traditional machine vision methods such as Speeded Up Robust Features (SURF) [4] significantly improved the model.”
MethodDeep Learning for Single-View Instance RecognitionPreprint
Held, D., Thrun, S., & Savarese, S. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1507.08286 · Cited by 27
Planetary Monocular Simultaneous Localization and Mapping
Bajpai, A., Burroughes, G., Shaukat, A., & Gao, Y. (2015) · Journal of Field Robotics · doi:10.1002/rob.21608 · Cited by 22
Saliency moments for image categorization
Реди, М., & Mérialdo, B. (2011) · International Conference on Multimedia Retrieval · doi:10.1145/1991996.1992035 · Cited by 26
Extracting salient region for pornographic image detection
Yan, C. C., Liu, Y., Xie, H., et al. (2014) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2014.03.005 · Cited by 21
Action recognition in still images by learning spatial interest regions from videos
Eweiwi, A., Cheema, M. S., & Bauckhage, C. (2014) · Pattern Recognition Letters · doi:10.1016/j.patrec.2014.07.017 · Cited by 21
Medical image registration based on fuzzy c-means clustering segmentation approach using SURF
Gupta, S., Chakarvarti, S. K., & Zaheeruddin, N. A. (2016) · International Journal of Biomedical Engineering and Technology · doi:10.1504/ijbet.2016.074113 · Cited by 21
“Bay et al. (2006) introduced a new algorithm called SURF.”
A compressive tracking based on time-space Kalman fusion model
Yun, X., Jing, Z., Xiao, G., Jin, B., & Zhang, C. (2015) · Science China Information Sciences · doi:10.1007/s11432-015-5356-0 · Cited by 22
“In the proposed approach, it is assumed that visible and infrared image sequences have been registered in time and space spans by some image registration approaches based on Walsh [30], SURF (speeded up robust features) [31], CSIFT (colored scale invariant feature transform) [32], and so on.”
MethodFrame buffer-less stream processor for accurate real-time interest point detection
Licciardo, G. D., Boesch, T., Pau, D. P., et al. (2016) · Integration · doi:10.1016/j.vlsi.2015.12.010 · Cited by 21
Unsupervised approach for object matching using Speeded Up Robust Features
Vardhan, A. H., Verma, N. K., Sevakula, R. K., et al. (2015) · IEEE Applied Imagery Pattern Recognition Workshop · doi:10.1109/aipr.2015.7444541 · Cited by 21
“[8] proposed a box filter approximation for this Gaussian partial derivative; thus making the operation less computationally complex while producing similar results.”
BackgroundResultInfluentialA Comparative Study of Similarity Measures for Content-Based Medical Image Retrieval.
Collins, J. J., & Okada, K. (2012) · Cited by 21
Towards comparison of underwater SLAM methods: An open dataset collection
Duarte, A. C., Zaffari, G. B., da Rosa, R. T. S., Longaray, L. M., Drews, P., & Botelho, S. S. D. C. (2016) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceans.2016.7761315 · Cited by 24
“When optical images are used, feature extraction are adopted to describe visual images such as Scale Invariant Feature Transform (SIFT) method [24] and the Speeded Up Robust Features (SURF) method [25].”
MethodPose-aware C-arm for automatic re-initialization of interventional 2D/3D image registration
Fotouhi, J., Fuerst, B., Johnson, A. A., Lee, S. C., Taylor, R. H., Osgood, G. M., Navab, N., & Armand, M. (2017) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-017-1611-8 · Cited by 24
“, SURF [1]) extracted from the color images and”
Predictive RANSAC: Effective model fitting and tracking approach under heavy noise and outliers
Li, Y., & Gans, N. (2017) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2017.05.013 · Cited by 21
Vision Based Pose Estimation of Multiple Peg-in-Hole for Robotic Assembly
Pitchandi, N., Perumaal, S. S., & Yogameena, B. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-68124-5_5 · Cited by 21
FPGA-Based Hardware Design for Scale-Invariant Feature Transform
Li, S., Wang, W., Pan, W.-Z., Hsu, C.-C. J., & Lu, C. (2018) · IEEE Access · doi:10.1109/access.2018.2863019 · Cited by 21
SeqSLAM++: View-based robot localization and navigation
Oishi, S., Inoue, Y., Miura, J., & Tanaka, S. (2018) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2018.10.014 · Cited by 25
“Local image descriptors such as SIFT [26] and SURF [27] are sometimes used for characterizing an image with, for example, the Bag-of-Visual-Words model [ 28].”
DOSeqSLAM: Dynamic On-line Sequence Based Loop Closure Detection Algorithm for SLAM
Tsintotas, K. A., Bampis, L., & Gasteratos, A. C. (2018) · IEEE International Conference on Imaging Systems and Techniques (IST) · doi:10.1109/ist.2018.8577113 · Cited by 25
“Following the extraction of SURF local features from the input camera data, SeqSLAM is applied.”
MethodUnderwater Place Recognition in Unknown Environments with Triplet Based Acoustic Image Retrieval
Ribeiro, P. O. C. S., dos Santos, M. M., Drews, P., Botelho, S. S. D. C., Longaray, L. M., Giacomo, G. G. D., & Pias, M. R. (2018) · International Conference on Machine Learning and Applications · doi:10.1109/icmla.2018.00084 · Cited by 21
Automated accurate registration method between UAV image and Google satellite map
Yuan, Y., Huang, W., Wang, X., et al. (2019) · Multimedia Tools and Applications · doi:10.1007/s11042-019-7729-7 · Cited by 21
A Lightweight Neural Network Based on AlexNet-SSD Model for Garbage Detection
Lee, S., Yeh, C.-H., Hou, T., et al. (2019) · doi:10.1145/3341069.3341087 · Cited by 21
“…on image eigenvalues includes BRIEF (Binary Description Very Fast) [6], BRISK (Binary Robust Invariant Scalable Keypoint) [7], FREAK (Fast Retina Keypoint ) [8], ORB (Oriented FAST and Rotated BRIEF) [9], SIFT (Scale-Invarient Feature Transform) [10] and SURF (Speeded Up Robust Feature) [11].”
BackgroundGraphite: Graph-Induced Feature Extraction for Point Cloud Registration
Saleh, M., Dehghani, S., Busam, B., Navab, N., & Tombari, F. (2020) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv50981.2020.00034 · Cited by 28
Motion Consistency-Based Correspondence Growing for Remote Sensing Image Matching
Liu, Y., Li, Y., Dai, L., et al. (2021) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2020.3048258 · Cited by 21
“…the following two steps [4]: 1) constructing initial putative matches based on the existing feature matching techniques, such as SIFT [5], SURF [6], and ORB [7]; Yanping Li is with the Department of Computer Science and Technology, Tongji University, Shanghai 201804, China (e-mail:…”
BackgroundDetection of COVID-19 in X-ray images by classification of bag of visual words using neural networks
Nabizadeh, Z., Karimi, N., Khadivi, P., Roshandel, R., Emami, A., & Samavi, S. (2021) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2021.102750 · Cited by 22
“SURF relies on the Hessian matrix ’ s determinant for selecting the location and the scale [27].”
InfluentialAppearance-Based Loop Closure Detection via Locality-Driven Accurate Motion Field Learning
Zhang, K., Jiang, X., & Ma, J. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3086822 · Cited by 22
Early detection of Alzheimer’s disease using local binary pattern and convolutional neural network
Francis, A., Pandian, S. I. A., & Initiative, T. A. D. N. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11161-y · Cited by 23
“Fast Hessian matrix [4] is employed to detect the significant voxels from the different slices of MRI images.”
MethodInfluentialRevisiting Binary Local Image Description for Resource Limited Devices
Suárez, I., Buenaposada, J. M., & Baumela, L. (2021) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2021.3107024 · Cited by 26
“SURF [24] approximates the first and second order derivatives with the integral image, that we also use to speed up our box differences.”
MegaStitch: Robust Large-Scale Image Stitching
Zarei, A., Gonzalez, E. M., Merchant, N. C., Pauli, D., Lyons, E. H., & Barnard, K. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3141907 · Cited by 19
Image Registration Techniques and Applications: Comparative Study on Remote Sensing Imagery
Ihmeida, M., & Wei, H. (2021) · International Conference on Developments in eSystems Engineering (DeSE) · doi:10.1109/dese54285.2021.9719538 · Cited by 21
Joint Optimization of Depth and Ego-Motion for Intelligent Autonomous Vehicles
Gao, Y., Tian, F., Li, J., Fang, Z., Al–Rubaye, S., Song, W., & Yan, Y. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2022.3159275 · Cited by 24
“These two types of methods construct a globally consistent pose and three-dimensional maps by tracking hand-crafted image geometric features (such as SIFT [7], SURF [8], ORB [9], and so on) over multiple frames, and by optimizing methods such as Bundle Adjustment or Kalman filter.”
MethodMultimodal-Based and Aesthetic-Guided Narrative Video Summarization
Xie, J., Chen, X., Zhang, T., Zhang, Y., Lu, S.-P., Cesar, P. S., & Yang, Y. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3183394 · Cited by 30
Intelligent Matching Method for Heterogeneous Remote Sensing Images Based on Style Transfer
Zhao, J., Yang, D., Li, Y., Xiao, P., & Yang, J. (2022) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2022.3197748 · Cited by 24
“The feature-based matching methods such as scale-invariant feature transform (SIFT) [2], speeded-up robust features (SURF) [3], and radiation-variation insensitive feature transform (RIFT) [4] achieve matching by extracting local invariant features such as points [5], lines [6], and surfaces [7] of images.”
MethodSelf-Supervised Rigid Registration for Multimodal Retinal Images
An, C., Wang, Y., Zhang, J., & Nguyen, T. Q. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3201476 · Cited by 26
Horticultural Image Feature Matching Algorithm Based on Improved ORB and LK Optical Flow
Chen, Q., Yao, L., Xu, L., Yang, Y., Xu, T., Yang, Y., & Liu, Y. (2022) · Remote Sensing · doi:10.3390/rs14184465 · Cited by 23
PHOTOGRAMMETRY NOW AND THEN – FROM HAND-CRAFTED TO DEEP-LEARNING TIE POINTS –
Morelli, L., Bellavia, F., Menna, F., & Remondino, F. (2022) · 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-xlviii-2-w1-2022-163-2022 · Cited by 33
3DG-STFM: 3D Geometric Guided Student-Teacher Feature Matching
Mao, R., Bai, C., An, Y., Zhu, F. M., & Lu, C. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19815-1_8 · Cited by 35
Liu, Y., Zhang, R., Deng, R., & Zhao, J. C. (2023) · GIScience & Remote Sensing · doi:10.1080/15481603.2023.2196159 · Cited by 24
Seamless UAV Hyperspectral Image Stitching Using Optimal Seamline Detection via Graph Cuts
Peng, Z., Ma, Y., Zhang, Y., Li, H., Fan, F., & Mei, X. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3275970 · Cited by 28
“The popular methods of feature matching are SIFT [16] and SURF [17].”
BackgroundMethodDarkFeat: Noise-Robust Feature Detector and Descriptor for Extremely Low-Light RAW Images
He, Y., Hu, Y., Wang, Z., Li, J., Liu, Y., Han, Y., & Wen, J. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i1.25161 · Cited by 28
“Early methods for keypoint and descriptor extraction use handcrafted features (Bay, Tuytelaars, and Gool 2006; Lowe 1999; Tardos 2016; Calonder et al. 2010; Rublee et al. 2011; Shi et al. 1994).”
MethodKanakis, M., Maurer, S., Spallanzani, M., Chhatkuli, A., & Gool, L. V. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00651 · Cited by 37
Self-citationBlock world reconstruction from spherical stereo image pairs
Kim, H., & Hilton, A. (2015) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2015.04.001 · Cited by 21
“We use SURF feature matching [49] between captured images for different stereo pairs.”
MethodContent Based Image Retrieval Using Bag-Of-Regions
Vieux, R., Benois‐Pineau, J., & Domenger, J. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-27355-1_47 · Cited by 21
Image Retrieval for Online Browsing in Large Image Collections
Mikulík, A., Chum, O., & Matas, J. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-41062-8_2 · Cited by 21
Binary Histogrammed Intensity Patches for Efficient and Robust Matching
Taylor, S., & Drummond, T. W. (2011) · International Journal of Computer Vision · doi:10.1007/s11263-011-0430-6 · Cited by 21
Routier‐Kierzkowska, A., & Kwiatkowska, D. (2008) · Functional Plant Biology · doi:10.1071/fp08047 · Cited by 22
“Subpixel feature points are first detected in the two images, with the aid of two different detectors, SURF (Bay et al. 2006) and SIFT (Lowe 2004), and thenmatched using the points descriptors introduced by Lowe (2004).”
MethodVerhoeven, G., Karel, W., Štuhec, S., et al. (2015) · 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-xl-5-w4-307-2015 · Cited by 21
“Although the input for feature extraction (i.e. IP detection and description) is generally a collection of true-colour frame images, the most common feature detectors and descriptors such as SIFT and SURF have been developed to work on single-band greyscale images, since this – amongst other reasons – greatly reduces…”
MethodInfluentialBackground subtraction adapted to PTZ cameras by keypoint density estimation
Guillot, C., Taron, M., Sayd, P., et al. (2010) · British Machine Vision Conference (BMVC) · doi:10.5244/c.24.34 · Cited by 21
Histogram equalization and image feature matching
Tu, L., & Dong, C. (2013) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp.2013.6744035 · Cited by 21
Kogler, M., & Lux, M. (2010) · doi:10.1145/1814245.1814248 · Cited by 21
“MDMKDD 10, July 25th, 2010 Washington, DC, USA Copyright 2010 ACM 978-1-4503-0220-3 ...$10.00. and [12], aims to tackle theses issues either with global features or with local feature like Scale Invariant Feature Transform (SIFT) [9] or Speeded Up Robust Features (SURF) [1].”
BackgroundInfluentialEfficient keyframe-based real-time camera tracking
Dong, Z., Zhang, G., Jia, J., & Bao, H. (2013) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2013.08.005 · Cited by 20
“Invariant feature descriptors [35] [4] require much less memory resource and can be used for global localization; but the computation is more expensive.. Chekhlov et al. [6] proposed using the appearance-based indexing on space and scale to facilitate the use of invariant feature descriptors for…”
Kumar, A. N., Miga, M. I., Pheiffer, T. S., Chambless, L. B., Thompson, R. C., & Dawant, B. M. (2014) · Medical Image Analysis · doi:10.1016/j.media.2014.07.004 · Cited by 23
“We use the SURF detector because of its fast computation time (Bay et al., 2008) to detect keypoints in the stereo-pair video streams.”
Explicit Occlusion Reasoning for 3D Object Detection
Meger, D., Wojek, C., Little, J. J., & Schiele, B. (2011) · British Machine Vision Conference (BMVC) · doi:10.5244/c.25.113 · Cited by 23
A Performance Evaluation of SIFT and SURF for Multispectral Image Matching
Saleem, S., Bais, A., & Sablatnig, R. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-31295-3_20 · Cited by 21
Panoramic video stitching in multi-camera surveillance system
He, B., Zhao, G., & Liu, Q. (2010) · International Conference on Image and Vision Computing New Zealand (IVCNZ) · doi:10.1109/ivcnz.2010.6148851 · Cited by 22
Image segmentation with one shape prior — A template-based formulation
Chen, S., Cremers, D., & Radke, R. J. (2012) · Image and Vision Computing · doi:10.1016/j.imavis.2012.09.005 · Cited by 21
“2 illustrates the SIFT features ([15]) detected on an ultrasound cardiac test image and a clean shape template.”
MethodInfluentialSuaib, N. M., Marhaban, M. H., Saripan, M. I., & Ahmad, S. A. (2014) · IEEE Region 10 Symposium (TENSYMP) · doi:10.1109/tenconspring.2014.6863025 · Cited by 22
“Due to the major drawback of SIFT which is relatively slow to compute, then Bay et al [12] presented an improved version of SIFT named Speeded Up Robust Feature that is much faster than SIFT.”
MethodMonocular SLAM for Autonomous Robots with Enhanced Features Initialization
Guerra, E., Munguía, R., & Grau, A. (2014) · Sensors · doi:10.3390/s140406317 · Cited by 23
“In the interest regions found, SURF-based feature descriptors [38] are matched to produce up to ten new stereo features to initialize the EKF state vector, see the end of the process at Figure 5.”
MethodRobotic grasping based on efficient tracking and visual servoing using local feature descriptors
Anh, T. L., & Song, J.-B. (2012) · International Journal of Precision Engineering and Manufacturing · doi:10.1007/s12541-012-0049-8 · Cited by 21
Fast Near-Duplicate Video Retrieval via Motion Time Series Matching
Zhang, J. R., Ren, J., Chang, F., Wood, T. L., & Kender, J. R. (2012) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2012.111 · Cited by 20
“We experimented with different local feature detectors including SIFT [10], SURF [11] and FAST [14].”
BackgroundMethodSalient region detection and segmentation for general object recognition and image understanding
Huang, T., Tian, Y., Li, J., & Yu, H. (2011) · Science China Information Sciences · doi:10.1007/s11432-011-4487-1 · Cited by 23
Autonomous navigation and environment modeling for MAVs in 3-D enclosed industrial environments
Li, Q., Li, D., Wu, Q., Tang, L., Huo, Y., Zhang, Y., & Cheng, N. (2013) · Computers in Industry · doi:10.1016/j.compind.2013.06.010 · Cited by 24
Vehicle matching in smart camera networks using image projection profiles at multiple instances
Jelača, V., Pižurica, A., Niño-Castañeda, J. O., et al. (2013) · Image and Vision Computing · doi:10.1016/j.imavis.2013.06.007 · Cited by 21
EVALUATION OF AREA-BASED IMAGE MATCHING APPLIED TO DTM GENERATION WITH HIRISE IMAGES
Re, C., Roncella, R., Forlani, G., et al. (2012) · ISPRS annals of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprsannals-i-4-209-2012 · Cited by 21
Automatic alignment of 3D reconstructions using a Digital Surface Model
Wendel, A., Irschara, A., & Bischof, H. (2011) · doi:10.1109/cvprw.2011.5981717 · Cited by 21
“Recent work of Val- gren and Lilienthal [20] investigates the performance of local features such as SIFT [12] and SURF [3] in scenes which undergo such changes.”
BackgroundTowards large-scale geometry indexing by feature selection
Tolias, G., Kalantidis, Y., Avrithis, Y., & Kollias, S. (2013) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2013.12.002 · Cited by 21
“In all experiments, we use SURF features and descriptors [42] and a kv = 100K generic visual codebook trained using approximate k-means (AKM) [2] on a set of images of urban scenes that are not part of our evaluation datasets.”
MethodA pure vision-based topological SLAM system
Lui, W. L. D., & Jarvis, R. A. (2012) · The International Journal of Robotics Research · doi:10.1177/0278364911435160 · Cited by 22
“For optimal performance, a Chow Liu tree which captures the co-occurrence statistics of visual words (SURF features (Bay et al. 2006)) was constructed from a large number of observations.”
MethodInfluentialTowards persistent indoor appearance-based localization, mapping and navigation using CAT-Graph
Maddern, W., Milford, M. J., & Wyeth, G. F. (2012) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2012.6386186 · Cited by 26
“Features are extracted using SURF [24], typically producing between 100 and 500 descriptors per frame.”
MethodA content-aware image stitching algorithm for mobile multimedia devices
Kim, H.-K., Lee, K.-W., Jung, J.-Y., et al. (2011) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2011.6131166 · Cited by 21
A Study of feature extraction algorithms for optical flow tracking
Nourani‐Vatani, N., Borges, P., & Roberts, J. M. (2012) · QUT ePrints (Queensland University of Technology) · Cited by 24
Towards Persistent Localization and Mapping with a Continuous Appearance-based Topology
Maddern, W., Milford, M. J., & Wyeth, G. F. (2012) · Robotics: Science and Systems · doi:10.15607/rss.2012.viii.036 · Cited by 20
“Feature descriptors are extracted using SURF [29], and a fast approximate [30] was used to find the corresponding visual word for each descriptor.”
MethodAutomatic Georeferencing of Aerial Images Using Stereo High-Resolution Satellite Images
Oh, J., Toth, C. K., & Grejner‐Brzezinska, D. A. (2011) · Photogrammetric Engineering & Remote Sensing · doi:10.14358/pers.77.11.1157 · Cited by 21
Hauswald, J., Laurenzano, M. A., Zhang, Y., Yang, H., Kang, Y., Li, C., Rovinski, A., Khurana, A., Dreslinski, R., Mudge, T., et al. (2016) · ACM Transactions on Computer Systems · doi:10.1145/2870631 · Cited by 19
“…for the user. et al. 2011], representing the state-of-the-art question-and-answer system based on IBM’s Watson [Ferrucci et al. 2010]; and SURF [Bay et al. 2006], implemented using OpenCV [Bradski 2000] and representing state-of-the-art image matching algorithms widely used in various…”
MethodInfluentialSaleem, S., Bais, A., & Sablatnig, R. (2016) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2016.05.005 · Cited by 21
Isolated sign language recognition using Improved Dense Trajectories
Özdemir, O., Camgöz, N. C., & Akarun, L. (2016) · Signal Processing and Communications Applications Conference (SIU) · doi:10.1109/siu.2016.7496151 · Cited by 19
“Kamera hareketinden kaynaklanan gezingeleri bulabilmek için, ardışıl çerçeveler arasında Hızlandırılmış Gürbüz Öznitelikler (Speeded Up Robust Features - SURF) [17] ve Rasgele Örneklemlerin Mutabakatı (Random Sample Consensus - RANSAC) [18] kullanılarak homografi hesaplanmaktadır.”
BORDER: An Oriented Rectangles Approach to Texture-Less Object Recognition
Chan, J., Lee, J. A., & Kemao, Q. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.312 · Cited by 22
Multi-gait recognition using hypergraph partition
Chen, X., Xu, J., & Weng, J. (2016) · Machine Vision and Applications · doi:10.1007/s00138-016-0810-6 · Cited by 21
Moving towards in object recognition with deep learning for autonomous driving applications
Uçar, A., Demir, Y., & Guzelis, C. (2016) · International Conference on Innovations in Intelligent Systems and Applications (INISTA) · doi:10.1109/inista.2016.7571862 · Cited by 24
Objects Classification by Learning-Based Visual Saliency Model and Convolutional Neural Network
Li, N., Zhao, X., Yang, Y., et al. (2016) · Computational Intelligence and Neuroscience · doi:10.1155/2016/7942501 · Cited by 21
Advanced Techniques in CBIR: Local Descriptors, Visual Dictionaries and Bags of Features
Junior, E. A. D. V., & Cord, M. (2009) · doi:10.1109/sibgrapi-tutorials.2009.14 · Cited by 24
On Pose Estimation in Room-Scaled Environments
Nyqvist, H. E. (2016) · Linköping University Electronic Press eBooks · doi:10.3384/lic.diva-132735 · Cited by 25
An automated CNN recommendation system for image classification tasks
Wang, S., Li, S., Fan, W., et al. (2017) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2017.8019347 · Cited by 21
Automatic parking identification and vehicle guidance with road awareness
Krasner, G., & Katz, E. (2016) · doi:10.1109/icsee.2016.7806133 · Cited by 21
“According to article [12], the ORB method was an order of magnitude faster than the SURF method and over two order faster than SIFT method, when using approximately the same processor and picture resolution as used in the presented system.”
MethodVisual domain adaptation using weighted subspace alignment
Chen, S., Zhou, F., & Liao, Q. (2016) · Visual Communications and Image Processing · doi:10.1109/vcip.2016.7805516 · Cited by 20
“Following the previous work [10], image representations are extracted as SURF features [15].”
MethodTHE EFFECT OF JPEG COMPRESSION IN CLOSE RANGE PHOTOGRAMMETRY
Akçay, Ö., Erenoğlu, R. C., & Avşar, E. Ö. (2017) · International Journal of Engineering and Geosciences · doi:10.26833/ijeg.287308 · Cited by 21
Zhang, H., Bai, J., Li, Z., et al. (2017) · Infrared Physics & Technology · doi:10.1016/j.infrared.2017.04.001 · Cited by 21
Feature points for multisensor images
Saleem, S., Bais, A., Sablatnig, R., Ahmad, A., & Naseer, N. (2017) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2017.04.032 · Cited by 21
A New Accurate and Fast Homography Computation Algorithm for Sports and Traffic Video Analysis
Liu, S., Chen, J., Chang, C.-H., & Ai, Y. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2017.2731781 · Cited by 22
BIND: Binary Integrated Net Descriptors for Texture-Less Object Recognition
Chan, J., Lee, J. A., & Kemao, Q. (2017) · 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.2017.322 · Cited by 21
A circular feature-based pose measurement method for metal part grasping
Wu, C., He, Z., Zhang, S., & Zhao, X. (2017) · Measurement Science and Technology · doi:10.1088/1361-6501/aa87ea · Cited by 21
Fast and robust copy-move forgery detection based on scale-space representation
Park, C., & Choeh, J. Y. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-5248-y · Cited by 20
CloudAR: A Cloud-based Framework for Mobile Augmented Reality
Zhang, W., Lin, S., Bijarbooneh, F. H., Cheng, H. F., & Hui, P. (2017) · Thematic Workshops of ACM Multimedia · doi:10.1145/3126686.3126739 · Cited by 26
Kokila, R., Sannidhan, M. S., & Bhandary, A. (2017) · International Conference on Advances in Computing, Communications and Informatics (ICACCI) · doi:10.1109/icacci.2017.8126046 · Cited by 21
“Here we used two robust feature descriptor, 1) SpeedUp Robust Feature (SURF) [8].”
Method
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.