SURF: Speeded Up Robust Features
Open accessThe DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index; none of the 18 sources its claims cite is checked for support yet.
Overview
CoChat Summary
TL;DR
SURF is a scale- and rotation-invariant interest point detector and descriptor that uses integral images, Hessian-based detection, and Haar-wavelet responses to be faster than prior methods while matching or exceeding their repeatability and distinctiveness. It reports strong results on standard benchmarks and a museum object recognition task.
Key findings
- The Fast-Hessian detector is more than 3× faster than DoG and 5× faster than Hessian-Laplace, with comparable or better repeatability on Graffiti, Leuven, Boat, and Wall; on the first Graffiti image (800×640), computation times were 120 ms for Fast-Hessian, 400 ms for DoG, 650 ms for Hessian-Laplace, and 1800 ms for Harris-Laplace.
- SURF’s 64-dimensional descriptor outperformed GLOH, SIFT, and PCA-SIFT in most comparisons, sometimes by more than 10% improvement in recall for the same precision; SURF-128 was slightly better but slower to match.
- Joint detector-descriptor computation times on the first Graffiti image were 255 ms for U-SURF, 354 ms for SURF, 391 ms for SURF-128, and 1036 ms for SIFT.
- In museum object recognition, recognition rates were 85.7% for SURF-128, 83.8% for U-SURF, 82.6% for SURF, 78.3% for GLOH, 78.1% for SIFT, and 72.3% for PCA-SIFT.
Methods
- Design: a scale- and rotation-invariant detector/descriptor. The detector uses the determinant of the Hessian with box-filter approximations evaluated via integral images; the descriptor uses Haar-wavelet responses in a 4×4 subregion grid, producing a 64-dimensional vector, with an upright U-SURF variant and an extended SURF-128 variant.
- Data: standard evaluation sequences from Mikolajczyk (including Graffiti, Wall, Boat, Leuven, Bikes, Trees, and Ubc; Bark excluded from descriptor evaluations) and a real-life museum object recognition database of 216 images of 22 objects, with 116 test images at 320×240.
- Analysis: detector repeatability scores, descriptor recall-(1-precision) graphs under similarity-threshold and nearest-neighbour-ratio matching, recognition rates, and computation timings on a standard Linux PC (Pentium IV, 3 GHz).
Limitations
- The paper has no explicit limitations section. It states that space limitations prevent showing results on all sequences, and it assumes skew, anisotropic scaling, and perspective effects are second-order effects covered only by overall descriptor robustness.
- Evident from the text: the method is scale- and rotation-invariant but not affine-invariant; Fig. 4 notes the interest points are not affine invariant, and the descriptor does not use colour.
- Evident from the text: the object recognition experiment uses small 320×240 test images and matching without additional geometric constraints, which the authors say would reduce false positives but were omitted for comparison purposes.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures








Verification
Claims the paper makes
0 Valid · 2 Not valid · 12 Not checked · 30 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The number of interest points found is on average very similar for all detectors.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The 4 × 4 sub-region division solution provided the best results.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“SURF outperformed the other descriptors for almost all the comparisons.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The SURF descriptor outperforms the other descriptors in a systematic and significant way, with sometimes more than 10% improvement in recall for the same level of precision.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As shown in the results section, the performance is comparable to the one using the discretised and cropped Gaussians.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
8 more claims
Create a free account to see all 44 claims and every source verdict.
- It may state two separate claims as one. 24
- Probably not a substantial claim. 13
- The quote only partly states this claim. 1
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“This transition marked a new era, where we moved beyond the classic approaches and task-specific feature engineering of earlier days [Lowe 2004; Bay et al. 2006; Rosten and Drummond 2006] towards models that could be trained once over large amounts of data, and then adapted for a broad variety of tasks, such as image recognition, object detection, and image segmentation [Krizhevsky et al. 2012; Sz”
From On the Opportunities and Risks of Foundation Models (2021) · cites this paper as Bay et al. 2006
Citations
Cites
Distinctive Image Features from Scale-Invariant Keypoints
Lowe, D. G. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000029664.99615.94 · Cited by 55,589
Rapid object detection using a boosted cascade of simple features
Viola, P., & Jones, M. (2005) · Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001 · doi:10.1109/cvpr.2001.990517 · Cited by 18,300
Object recognition from local scale-invariant features
Lowe, D. G. (1999) · Proceedings of the Seventh IEEE International Conference on Computer Vision · doi:10.1109/iccv.1999.790410 · Cited by 16,472
A Combined Corner and Edge Detector
Harris, C., & Stephens, M. (1988) · Procedings of the Alvey Vision Conference 1988 · doi:10.5244/c.2.23 · Cited by 12,901
A performance evaluation of local descriptors
Mikolajczyk, K., & Schmid, C. (2005) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2005.188 · Cited by 6,723 · cited 14×
“This PCA-SIFT yields a 36-dimensional descriptor which is fast for matching, but proved to be less distinctive than SIFT in a second comparative study by Mikolajczyk et al. [8] and slower feature computation reduces the effect of fast matching.”
BackgroundMethodResultInfluentialMarr, D., & Hildreth, E. (1980) · Proceedings of the Royal Society of London. Series B. Biological Sciences · doi:10.1098/rspb.1980.0020 · Cited by 6,510
PCA-SIFT: a more distinctive representation for local image descriptors
Ke, Y., & Sukthankar, R. (2004) · Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004. · doi:10.1109/cvpr.2004.183 · Cited by 3,932
Robust wide-baseline stereo from maximally stable extremal regions
Matas, J., Chum, O., Urban, M., & Pajdla, T. (2004) · Image and Vision Computing · doi:10.1016/j.imavis.2004.02.006 · Cited by 3,686
Scalable Recognition with a Vocabulary Tree
Nister, D., & Stewenius, H. (2006) · 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2 (CVPR'06) · doi:10.1109/cvpr.2006.264 · Cited by 3,579
Scale & Affine Invariant Interest Point Detectors
Mikolajczyk, K., & Schmid, C. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000027790.02288.f2 · Cited by 3,436
Cited by
Items 6,101–6,200 of 14,752
Research on Video Image Stitching Technology Based on SURF
Zhang, Y., Yang, L., & Wang, Z. (2012) · International Symposium on Computational Intelligence and Design · doi:10.1109/iscid.2012.235 · Cited by 8
Blighe, M., & O'Connor, N. E. (2008) · ACM International Conference on Image and Video Retrieval (CIVR) · doi:10.1145/1386352.1386382 · Cited by 9
Augmented Reality for maintenance operator training using SURF points and homography
Paz, A., Guenaga, M., & Eguíluz, A. (2012) · International Conference on Remote Engineering and Virtual Instrumentation · doi:10.1109/rev.2012.6293130 · Cited by 8
Real-Time Image Mosaicing Using Non-rigid Registration
de Souza, R. H. C., Okutomi, M., & Torii, A. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-25367-6_28 · Cited by 8
“To do so, the following algorithm is proposed: (1) the features in both frames being compared are detected using SURF descriptors [1]); (2) the nearest-neighbor matching of the features is computed; (3) a histogram of the matched descriptors is computed; (4) the overlap measure (OM) is computed.”
MethodSelf help: Seeking out perplexing images for ever improving navigation
Paul, R., & Newman, P. M. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5980404 · Cited by 9
Nicosevici, T., García, R., Negahdaripour, S., et al. (2007) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/robot.2007.364245 · Cited by 8
“The problem of extracting key image features has represented a topic of intensive research in the last decade, resulting in the development of a variety of interest point detectors: Shi and Tomasi [7], SIFT [8], SURF [9], affine covariant [10], etc.”
MethodEdge-Based Template Matching with a Harmonic Deformation Model
Hofhauser, A., Steger, C. T., & Navab, N. (2009) · Communications in computer and information science · doi:10.1007/978-3-642-10226-4_14 · Cited by 8
Efficient large scale SLAM including data association using the Combined Filter
Lerma, C. D. C., Ramos, F. T., & Neira, J. (2009) · Cited by 8
Image Processing Methods Applied to Landmine Detection in Ground Penetrating Radar
Sakaguchi, R. T. (2013) · DukeSpace (Duke University) · Cited by 12
Usui, H., Subramanian, L., Chang, K. K., et al. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1505.07502 · Cited by 8
Sparse Graphical Representation based Discriminant Analysis for Heterogeneous Face RecognitionPreprint
Peng, C., Gao, X., Wang, N., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1607.00137 · Cited by 8
Applying information theory to efficient SLAM
Chli, M. (2009) · Spiral (Imperial College London) · doi:10.25560/5634 · Cited by 18
Line Primitives and Their Applications in Geometric Computer Vision
Zhang, L. (2013) · MACAU: Open Access Repository of Kiel University (University Library Kiel) · Cited by 8
UAV Visual Teach and Repeat Using Only Semantic Object FeaturesPreprint
Toudeshki, A. G., Shamshirdar, F., & Vaughan, R. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1801.07899 · Cited by 8
Visual SLAM with Network Uncertainty Informed Feature Selection.Preprint
Ganti, P., & Waslander, S. L. (2018) · arXiv (Cornell University) · Cited by 8
Phillips, S., & Daniilidis, K. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1901.02078 · Cited by 8
Recurrent Attention Models for Depth-Based Person IdentificationPreprint
Haque, A., Alahi, A., & Fei-Fei, L. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.07212 · Cited by 8
Object-Centric Representation Learning from Unlabeled VideosPreprint
Gao, R., Jayaraman, D., & Grauman, K. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1612.00500 · Cited by 8
Self-Supervised 3D Keypoint Learning for Ego-motion Estimation
Tang, J., Ambrus, R., Guizilini, V., Pillai, S., Kim, H., Jensfelt, P., & Gaidon, A. (2019) · arXiv · doi:10.48550/arxiv.1912.03426 · Cited by 14
Towards Better Generalization: Joint Depth-Pose Learning without PoseNetPreprint
Zhao, W., Liu, S., Shu, Y., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.01314 · Cited by 8
S2DNet: Learning Accurate Correspondences for Sparse-to-Dense Feature MatchingPreprint
Germain, H., Bourmaud, G., & Lepetit, V. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.01673 · Cited by 19
“To perform keypoint detection, a variety of both hand-crafted [7,18,25,27] and learning-based [32,62] detectors have been developed, each aiming to detect accurate keypoints in both a repeatable as well as illumination, scale and affine invariant fashion.”
MethodRecent Advances in 3D Object and Hand Pose EstimationPreprint
Lepetit, V. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.05927 · Cited by 21
Temporal Context Aggregation for Video Retrieval with Contrastive LearningPreprint
Shao, J., Wen, X., Zhao, B., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.01334 · Cited by 8
“…employed handcrafted features including the Scale-Invariant Feature Trans-form (SIFT) features [29, 42, 72], the Speeded-Up Robust Features (SURF) [5, 10], Colour Histograms in HSV space [17, 30, 60], and Local Binary Patterns (LBP) [56, 73, 76], etc. Deep Convolutional Neural Networks (CNNs)…”
MethodUnsupervisedR&R: Unsupervised Point Cloud Registration via Differentiable RenderingPreprint
Banani, M. E., Gao, L., & Johnson, J. (2021) · arXiv (Cornell University) · Cited by 9
Benchmarking of Computer Vision Algorithms for Driver Monitoring on Automotive-grade Devices
Battiato, S., Conoci, S., Leotta, R., Ortis, A., Rundo, F., & Trenta, F. (2020) · 2020 AEIT International Conference of Electrical and Electronic Technologies for Automotive (AEIT AUTOMOTIVE) · doi:10.23919/aeitautomotive50086.2020.9307437 · Cited by 7
“Specifically, the authors implemented a robust and adaptive skin segmentation method to detect facial regions (e.g., eyes, lips, nose, eyebrows, etc.) by extracting Speeded Up Robust Features (SURF) facial features [13].”
BackgroundChang, Y., Tian, Y., How, J. P., & Carlone, L. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9561090 · Cited by 115
The effect of rainfall on feature points extraction and image stitching
Chia, W. C., Yeong, L. S., Ch'ng, S. I., Seng, K. P., & Ang, L.-M. (2014) · 2014 International Conference on Information Science, Electronics and Electrical Engineering · doi:10.1109/infoseee.2014.6946146 · Cited by 7
“This includes the Harris corner detector [1], the ShiftInvariant Feature Transform (SIFT) detector [2], and the Speeded-Up Robust Feature (SURF) detector [3].”
MethodHotFoot: Foot-Based User Identification Using Thermal Imaging
Saad, A., Izadi, K., Khan, A. A., Knierim, P., Schneegaß, S., Alt, F., & Abdelrahman, Y. (2023) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3544548.3580924 · Cited by 13
Arshad, S., & Park, T.-H. (2024) · Sensors · doi:10.3390/s24030906 · Cited by 6
Wide baseline stereovision based obstacle detection for unmanned surface vehicles
Jin, J., Liu, D., Li, F., et al. (2024) · Signal Image and Video Processing · doi:10.1007/s11760-024-03098-0 · Cited by 7
Yuan, W., Poosa, S. R. P., & Dirks, R. F. (2024) · Journal of Imaging · doi:10.3390/jimaging10050105 · Cited by 11
Tan, Y., Ji, P., Zhang, Y., Ge, F., & Zhu, S. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109425 · Cited by 8
He, R., Long, S., Sun, W. J., & Liu, H. (2024) · Drones · doi:10.3390/drones8110651 · Cited by 11
“The SURF algo-rithm [13], for instance, significantly improved execution efficiency using integral images on the Hessian matrix and reduced-dimensional feature descriptors, addressing the high computational complexity and time consumption of SIFT.”
Swin-transformer for weak feature matching
Guo, Y. C., Li, W., & Zhai, P. (2025) · Scientific Reports · doi:10.1038/s41598-025-87309-9 · Cited by 7
FPGA architecture-based front-end processing for SLAM applications
Bouazzaoui, I. E., Rodríguez, S., & Elouardi, A. (2025) · Journal of Real-Time Image Processing · doi:10.1007/s11554-025-01650-w · Cited by 7
“Several feature extractors are used in various SLAM systems, such as ORB [1], SIFT [2], and SURF [3].”
Shen, Y., Zhou, S., Fu, J., Wang, R., Chen, S., & Zheng, N. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3556859 · Cited by 13
“In early VPR systems [11], [12], the hand-crafted local features such as SIFT [13], SURF [14] and ORB [15] have been widely used to detect and describe keypoints, while they cannot handle severe appearance changes.”
Stereo-point tracking of inherent structural features for 3D computer vision measurements
Bottalico, F., & Sabato, A. (2025) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2025.112937 · Cited by 8
Jabbar, M. S., Siddique, T. H. M., Huang, K., & Khan, S. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.114325 · Cited by 14
A Review of AI in Human-Machine Cooperation: Machine Perspective
Sourav, M. S. G., Javaid, A. Y., & Cheng, L. (2025) · ACM Transactions on Autonomous and Adaptive Systems · doi:10.1145/3774318 · Cited by 13
“Human gestures can be identified [147] using machine learning through hand crafted features [137], [153], [19], [197], [172], [223] learnt by machine learning approaches [220].”
Evaluating the Fuzzy Coverage Model for 3D Multi-camera Network Applications
Mavrinac, A., Herrera, J. L. A., & Chen, X. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-16584-9_66 · Cited by 7
Zhang, S. S., Wang, X., Lin, Y., et al. (2015) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2015.2478055 · Cited by 7
“Local descriptor [25], [4] has been widely used in current image retrieval systems [17], [29].”
BackgroundTransfer learning using adaptive SVM for image classification
Jain, A., Srivastava, S., & Soman, S. (2013) · International Conference on Image Information Processing (ICIIP) · doi:10.1109/iciip.2013.6707659 · Cited by 7
Localization and stabilization of micro aerial vehicles based on visual features tracking
Chudoba, J., Saska, M., Báča, T., et al. (2014) · International Conference on Unmanned Aircraft Systems (ICUAS) · doi:10.1109/icuas.2014.6842304 · Cited by 7
Catadioptric omnidirectional image inpainting via a multi-scale approach and image unwrapping
Paredes, D. V., Rodríguez, P., & Ragot, N. (2013) · IEEE International Symposium on Robotic and Sensors Environments · doi:10.1109/rose.2013.6698420 · Cited by 7
“However, Liu ,in [14], proposed to use also gradient information as aditional descriptor (SURF - descriptors [16]) while working with a multi-scale approach which improve the overall performance.”
BackgroundA new affine invariant method for image matching
Palomares, J.-L., Montesinos, P., & Diep, D. (2012) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.910219 · Cited by 7
Distributed object recognition via feature unmixing
Luo, J., & Qi, H. (2010) · ACM/IEEE International Conference on Distributed Smart Cameras (ICDSC) · doi:10.1145/1865987.1865999 · Cited by 7
Image Copy-Move Forgery Detection Based on SIFT and Gray Level
Shen, X. J., Zhu, Y. H., Lv, Y. D., et al. (2012) · Applied Mechanics and Materials · doi:10.4028/www.scientific.net/amm.263-266.3021 · Cited by 7
“Bay presented SURF (Speeded-Up Robust Features) algorithm [10].”
MethodEvaluation Method, Dataset Size or Dataset Content: How to Evaluate Algorithms for Image Matching?
Kanwal, N., Bostancı, E., & Clark, A. F. (2016) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-015-0626-4 · Cited by 7
Local Descriptors without Orientation Normalization to Enhance Landmark Regconition
Truong, D.-D., Ngoc, C.-S. N., Nguyen, V.-T., et al. (2013) · Advances in intelligent systems and computing · doi:10.1007/978-3-319-02741-8_34 · Cited by 7
A vision-based automatic hotline tracking approach using unscented Kalman filter
Valipour, S., & Moradi, H. (2014) · doi:10.1109/icrom.2014.6990969 · Cited by 7
Fast Covariant VLAD for Image Search
Zhao, W., Ngo, C., & Wang, H. (2016) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2016.2585023 · Cited by 9
Virtual reality backend for operator controlled nanomanipulation
Sauvet, B., Ouarti, N., Haliyo, S. D., et al. (2012) · International Conference on Manipulation, Manufacturing and Measurement on the Nanoscale (3M-NANO) · doi:10.1109/3m-nano.2012.6472955 · Cited by 7
“Speeded Up Robust Features (SURF) is a technique that extracts robust features in an image [16].”
MethodMultiview Feature Analysis via Structured Sparsity and Shared Subspace Discovery
Chang, Y., Nie, F., & Ming-yu, W. (2017) · Neural Computation · doi:10.1162/neco_a_00977 · Cited by 6
“For example, to describe visual objects in images, histogram of oriented gradients (HOG) (Dalal & Triggs, 2005), speeded-up robust features (SURF) (Bay, Tuytelaars, & Gool, 2006), and scale-invariant feature transform (SIFT) (Lowe, 2004) have been widely used to extract lo-cal visual features.”
MethodJiang, D., & Kim, J. (2018) · Applied Sciences · doi:10.3390/app8101735 · Cited by 7
Secure Feature Extraction in Computational Vision Using Fully Homomorphic Encryption
Shortell, T., & Shokoufandeh, A. (2018) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-02683-7_15 · Cited by 7
Line Segments based Rotation Invariant Descriptor for Disparate Images
Sharma, T., Agrawal, P., Sahoo, P., et al. (2018) · IEEE Applied Imagery Pattern Recognition Workshop · doi:10.1109/aipr.2018.8707401 · Cited by 7
“Some examples of such implementations are discussed in [11]– [14] using SURF features.”
BackgroundInfluentialA Feature-Based Hybrid Medical Image Watermarking Algorithm Based on SURF-DCT
Nawaz, S. A., Li, J., Liu, J., et al. (2019) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-32591-6_118 · Cited by 7
Hosseini, K., Ebadi, H., & Ahmadi, F. F. (2020) · Journal of the Indian Society of Remote Sensing · doi:10.1007/s12524-020-01187-4 · Cited by 7
“In (Samadzadegan and Abdi 2012; Wang et al. 2012) the SIFT (speeded-up robust features) algorithm (Lowe 2004) and in (Samadzade-gan and Abdi 2011) the SURF (scale-invariant feature transform) algorithm (Bay et al. 2006) are used for matching and determining UAV location.”
MethodContent-based image retrieval for categorized dataset by aggregating gradient and texture features
Srivastava, D., Rajitha, B., & Agarwal, S. (2021) · Neural Computing and Applications · doi:10.1007/s00521-020-05614-y · Cited by 9
“The approaches [37, 38] use SURF combined with texture for the classification and retrieval of images having a single object.”
BackgroundInfluentialÜberatlas: Robust Speed-Up of Feature-Based Registration and Multi-Atlas Segmentation
Alvén, J., Norlén, A., Enqvist, O., & Kahl, F. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-19665-7_8 · Cited by 8
EfficientUNet: Modified encoder‐decoder architecture for the lung segmentation in chest x‐ray images
Agrawal, T., & Choudhary, P. (2022) · Expert Systems · doi:10.1111/exsy.13012 · Cited by 10
“Scale Invariant Feature Transform (Lowe, 2004) and Speeded-Up Robust Features (Bay et al., 2006) are most famous algorithms that are used for feature detection in handcraft designing.”
MethodA High-Resolution Remote Sensing Image Registration Method Combining Object and Point Features
Lu, J., Hu, Q., Zhu, R., & Jia, H. (2022) · IEEE Transactions on Aerospace and Electronic Systems · doi:10.1109/taes.2022.3159320 · Cited by 7
“5) Refinement Stage With Different PFB Methods: SIFT, SURF, and ORB are performed as the PFB methods of the refinement stage, respectively, for comparison, and the corresponding experimental results are described in Tables III–V. Experiments reveal that, among the three, SIFT has the best accuracy and most NOCC, while O…”
BackgroundKostrykin, L., & Rohr, K. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3185583 · Cited by 11
Sabri, N., Hamed, H. N. A., Ibrahim, Z., Ibrahim, K., & Isa, M. A. (2022) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2022.029221 · Cited by 7
Automatic marker-free registration of single tree point-cloud data based on rotating projection
Xu, X., Wang, P., Gan, X., et al. (2022) · Artificial Intelligence in Agriculture · doi:10.1016/j.aiia.2022.09.005 · Cited by 7
Underwater visual mapping of curved ship hull surface using stereo vision
Chung, D., & Kim, J. (2022) · Autonomous Robots · doi:10.1007/s10514-022-10071-8 · Cited by 7
Multi‐scale homography estimation based on dual feature aggregation transformer
Li, Y., Chen, K., Sun, S., & He, C. (2022) · IET Image Processing · doi:10.1049/ipr2.12722 · Cited by 9
Global Homography Motion Compensation for Versatile Video Coding
Li, Y., Li, Z., Li, L., et al. (2022) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip56404.2022.10008901 · Cited by 7
“In this work, SURF [10] extraction and matching are employed to generate the point correspondences.”
MethodAasha Nandhini, S., Hemalatha, R., Radha, S., et al. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15442-6 · Cited by 7
Method for the automatic recognition of cropland headland images based on deep learning
Qiao, Y., Liu, H., Meng, Z., Chen, J., & Ma, L. (2023) · International journal of agricultural and biological engineering · doi:10.25165/j.ijabe.20231602.6195 · Cited by 8
Multiview Structural Large Margin Classifier and its Safe Acceleration Strategy
Zhao, J., & Xu, Y. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3290540 · Cited by 8
“5 We manually extract or use preextracted feature representations, and the statistics of all datasets are summarized in Table I Six preextracted feature representations are adopted, that is, Global Color Histogram (CQ), LSS [44], PHOG [45], SIFT [46], RGSIFT [47], and SURF [48].”
MethodZhou, Q., Shen, S., Yu, S., et al. (2023) · The Visual Computer · doi:10.1007/s00371-023-03033-3 · Cited by 7
“In this paper, an image geometric transform method based on Speeded Up Robust Features (SURF) [46] and Maximum Likelihood Estimator Sample Consensus (MLESAC) [47] algorithmisproposedtousedforimagegeometriccorrection.”
Self-supervised Interest Point Detection and Description for Fisheye and Perspective Images
Mera-Trujillo, M., Patel, S., Gu, Y., et al. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00691 · Cited by 7
An Improved Visual SLAM Method with Adaptive Feature Extraction
Guo, X., Lyu, M., Xia, B., et al. (2023) · Applied Sciences · doi:10.3390/app131810038 · Cited by 7
Deep Semantic Feature Reduction for Efficient Remote Sensing Image Retrieval
Yelchuri, R., Khadidos, A. O., Khadidos, A. O., Alshareef, A. M., Swain, G., & Dash, J. K. (2023) · IEEE Access · doi:10.1109/access.2023.3324133 · Cited by 7
Unsupervised Image Sequence Registration and Enhancement for Infrared Small Target Detection
Hou, R., Yan, P., Duan, X., et al. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3392307 · Cited by 7
“…(SIFT) [23], speeded-up robust features (SURF) [24], and oriented FAST and rotated BRIEF (ORB) [25]), feature description (SIFT [23], SURF [24], and ORB [25]), feature matching (fast library for approximate nearest neighbors (FLANN) [26]), and final estimation ([27], random sample…”
BackgroundKostrzewa, A., Farella, E. M., Morelli, L., Ostrowski, W., Remondino, F., & Bakuła, K. (2024) · Applied Sciences · doi:10.3390/app14093635 · Cited by 10
Computer Vision and Augmented Reality for Human-Centered Fatigue Crack Inspection
Mojidra, R., Li, J., Mohammadkhorasani, A., et al. (2024) · Sensors · doi:10.3390/s24113685 · Cited by 7
“Notable algorithms for feature point detection include Harris–Stephens [21], Scale Invariant Feature Transform (SIFT) [22], Speeded-Up Robust Features (SURF) [23], Shi–Tomasi [24], and Features from Accelerated Segment Test (FAST) [25].”
MethodEnhancing CNN model classification performance through RGB angle rotation method
Doğan, Y., Özdemir, C., & Kaya, Y. (2024) · Neural Computing and Applications · doi:10.1007/s00521-024-10232-z · Cited by 11
“Over several decades, researchers have developed a multitude of hand-crafted features, such as histogram of oriented gradients (HOGs) [1], local binary patterns (LBPs) [2], haar-like features [3], scale-invariant feature transform (SIFT) [4], and speeded-up robust features (SURFs) [5].”
CLG-Net: Rethinking Local and Global Perception in Lightweight Two-View Correspondence Learning
Shen, M., Xiao, G., Yang, C., Guo, J., & Zhu, L. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3457816 · Cited by 9
“Conventional methods rely on feature detectors [5], [6], [7] and nearest neighbor algorithms [8] to establish initial correspondences, which inevitably experiences a significant amount of mismatches due to the possible challenging image variants, e . g .”
A SURF and SVD-based robust zero-watermarking for medical image integrity
Taj, R., Feng, T., Kanwal, S., et al. (2024) · PLoS ONE · doi:10.1371/journal.pone.0307619 · Cited by 7
Raising the Ceiling: Conflict-Free Local Feature Matching with Dynamic View Switching
Lu, X., & Ду, С. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72946-1_15 · Cited by 11
Precision in Safety: YOLOv9 vs. YOLOv10 for Helmet Image Detection
A’la, F. Y., Firdaus, N., Hartatik, H., et al. (2024) · doi:10.1109/ivit62102.2024.10692595 · Cited by 7
Towards Long Term SLAM on Thermal Imagery
Keil, C., Gupta, A., Kaveti, P., & Singh, H. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802139 · Cited by 11
“ORB [26], SIFT [20], and SURF [1] are foundational hand-tuned algorithms that have shaped the development of feature extraction and matching in computer vision.”
CRPlace: Camera-Radar Fusion with BEV Representation for Place Recognition
Fu, S., Duan, Y., Li, Y., Meng, C., Wang, Y., Ji, J., & Zhang, Y. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802267 · Cited by 7
Thomas, A., Kinnari, J., Lusk, P. C., Kondo, K., & How, J. P. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10801471 · Cited by 14
“In feature-based SLAM, the environment is described by consistently detectable features within a series of images like ORB [7], SIFT [8], SURF [9] or learned features [10], [11].”
Multi-channel Shape-Flow Kernel Descriptors for Robust Video Event Detection and Retrieval
Natarajan, P., Wu, S., Vitaladevuni, S. N. P., et al. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33709-3_22 · Cited by 7
Automated Annotation of Landmark Images Using Community Contributed Datasets and Web Resources
Jones, G. J. F., Byrne, D., Hughes, M. M., O'Connor, N. E., & Salway, A. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-23017-2_8 · Cited by 7
“We selected a combination of spatial data, low-level image features and SURF local image features [4].”
Towards the Optimal Hardware Architecture for Computer Vision
Nieto, A., L√≥pez, D., & Brea, V. (2012) · InTech eBooks · doi:10.5772/34023 · Cited by 7
A Parallel Analysis on Scale Invariant Feature Transform (SIFT) Algorithm
Yang, D., Liu, L., Zhu, F., & Zhang, W. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-24151-2_8 · Cited by 9
Synthetic Evidential Study as Primordial Soup of Conversation
Nishida, T., Nakazawa, A., Ohmoto, Y., Nitschke, C., Mohammad, Y. F. O., Thovuttikul, S., Lala, D., Abe, M., & Ookaki, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16313-0_6 · Cited by 7
Do humans fixate on interest points?
Dave, A., Dubey, R., & Ghanem, B. (2012) · DR-NTU (Nanyang Technological University) · Cited by 7
Maximum Entropy Threshold Segmentation for Target Matching Using Speeded-Up Robust Features
Zhou, M., Hong, X., Tian, Z., et al. (2014) · Journal of Electrical and Computer Engineering · doi:10.1155/2014/768519 · Cited by 7
“By using the integral images for the image convolution, SURF only requires a small number of histograms to quantize the gradient orientations [17].”
MethodSVM-based fast pedestrian recognition using a hierarchical codebook of local features
Besbes, B., Labbé, B., Rogozan, A., et al. (2010) · IEEE International Workshop on Machine Learning for Signal Processing (MLSP) · doi:10.1109/mlsp.2010.5589005 · Cited by 7
Dzulfahmi, & Ohta, N. (2013) · IAPR Asian Conference on Pattern Recognition (ACPR) · doi:10.1109/acpr.2013.159 · Cited by 7
“Bay [7] contributed on a faster and efficient computation in extracting the feature point by introducing SURF.”
MethodMarkerless tracking using interest window for augmented reality applications
Hamidia, M., Zenati-Henda, N., Belghit, H., et al. (2014) · doi:10.1109/icmcs.2014.6911133 · Cited by 7
New Sensors for Cultural Heritage Metric Survey: The ToF Cameras
Chiabrando, F., Piatti, D., & Rinaudo, F. (2011) · Geoinformatics FCE CTU · doi:10.14311/gi.6.38 · Cited by 7
Mobile camera based cross-screen interaction by object matching and tracking
Hu, W., Lian, S., Song, X., et al. (2013) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2013.6626224 · Cited by 7
“The SURF matching is adopted here for its better computational efficiency than SIFT, as well as good matching accuracy.”
MethodInfluentialCasaburi, L., Colace, F., Santo, M. D., & Greco, L. (2015) · Journal of Visual Languages & Computing · doi:10.1016/j.jvlc.2015.01.001 · Cited by 7
Straub, J. (2014) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2050172 · Cited by 7
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.