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 7,101–7,200 of 14,752
Person re-identification based on saliency
Wang, C., Tang, S. Y., Zhu, S., et al. (2016) · Chinese Control Conference (CCC) · doi:10.1109/chicc.2016.7553958 · Cited by 5
Ahn, J., Yun, I. S., Yoo, H., et al. (2016) · Eye · doi:10.1038/eye.2016.204 · Cited by 5
Moving Object Detection for Driving Assistance System Based on Improved ORB Feature Matching
Jun, G., & Zhu, H. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-45940-0_41 · Cited by 5
Video-based fire detection with spatio-temporal SURF and color features
Shi, L., Long, F., Zhan, Y., & Lin, C.-H. (2016) · World Congress on Intelligent Control and Automation · doi:10.1109/wcica.2016.7578269 · Cited by 6
Han, X., Li, G., Lin, W., et al. (2012) · Cited by 5
Subregion mosaicking applied to nonideal iris recognition
Yang, T., Stahl, J. S., Schuckers, S., & Hua, F. (2014) · IEEE Symposium on Computational Intelligence in Biometrics and Identity Management (CIBIM) · doi:10.1109/cibim.2014.7015455 · Cited by 5
Object Extraction and Classification in Video Surveillance Applications
Civelek, M., & Yazıcı, A. (2016) · European Review · doi:10.1017/s1062798716000582 · Cited by 5
“Speeded Up Robust Features (SURF) is a similar approach proposed in the study of Bay et al ., 29 which claims to be more ef fi cient in terms of feature extraction.”
MethodMonocular visual object-localization using natural corners for assembly tasks
Gu, J., Wang, H., Chen, W., et al. (2016) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio.2016.7866520 · Cited by 5
A 700fps Optimized Coarse-to-Fine Shape Searching Based Hardware Accelerator for Face Alignment
Wang, Q., Liu, L., Zhu, W., Mo, H., Deng, C., & Wei, S. (2017) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/3061639.3062182 · Cited by 5
“LD-SURF shows better performance than SIFT.”
MethodInfluentialMin, H., Zhao, X., Xu, Z., & Zhang, L. (2016) · KSII Transactions on Internet and Information Systems · doi:10.3837/tiis.2017.01.016 · Cited by 5
Combining semantics and augmented reality to support the human mind
Michahelles, F., Ciortea, A., García, K., et al. (2017) · doi:10.1145/3123024.3129270 · Cited by 5
A fast image retrieval method with convolutional neural networks
Yao, J., Deng, Y., Yao, Y., et al. (2017) · Chinese Control Conference (CCC) · doi:10.23919/chicc.2017.8029131 · Cited by 5
“In Euclidean space, the Euclidean distances were computed to determine the similarity of the two pictures [3].”
MethodRobust Relocalization Based on Active Loop Closure for Real-Time Monocular SLAM
Chen, X., Lu, H., Xiao, J., Zhang, H., & Wang, P. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-68345-4_12 · Cited by 5
Automatic measurement based on stereo vision system using a single PTZ camera
Wang, R., & Feng, S. (2017) · IEEE International Conference on Imaging Systems and Techniques (IST) · doi:10.1109/ist.2017.8261469 · Cited by 5
“Due to the use of box filters and integral images [11], we apply box filters of any size at exactly the same speed directly on the original image”
MethodA common feature-based disparity control strategy in stereoscopic panorama generation
Wang, H., Sandin, D. J., & Schonfeld, D. (2017) · IEEE Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip.2017.8305051 · Cited by 5
“, SIFT or SURF [11], [12]) from the four images as d1, d2, d3, d4, respectively.”
MethodMethod of Blob detection based on radon transform
Chen, C., Chen, Q., Gao, C., et al. (2018) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc.2018.8408138 · Cited by 5
Person Identification in Smart Surveillance Robots using Sparse Interest Points
Vinay, A., Sai Krishna, B., V, Manoj, P., et al. (2018) · Procedia Computer Science · doi:10.1016/j.procs.2018.07.114 · Cited by 5
Multiple Modal Features and Multiple Kernel Learning for Human Daily Activity Recognition
Vo, V. H., & Pham, H. M. (2018) · Science and Technology Development Journal · doi:10.32508/stdj.v21i2.441 · Cited by 5
Siekański, P., Michoński, J., Bunsch, E., & Sitnik, R. (2018) · ISPRS International Journal of Geo-Information · doi:10.3390/ijgi7120479 · Cited by 6
Xipnitou, M., Soile, S., Tziranis, I., et al. (2019) · Communications in computer and information science · doi:10.1007/978-3-030-12957-6_24 · Cited by 5
MOUNTAINOUS REMOTE SENSING IMAGES REGISTRATION BASED ON IMPROVED OPTICAL FLOW ESTIMATION
Feng, R., Li, X., & Shen, H. (2019) · ISPRS annals of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-annals-iv-2-w5-479-2019 · Cited by 5
Fotokapan Görüntülerinde Yerel Öznitelikler ile Nesne Tespiti
Şımşek, E., Özyer, B., & Özyer, G. T. (2019) · Gümüşhane Üniversitesi Fen Bilimleri Enstitüsü Dergisi · doi:10.17714/gumusfenbil.510717 · Cited by 5
Rahmat, R. F., Gunawan, D., Faza, S., et al. (2018) · International Conference on Informatics and Computing (ICIC) · doi:10.1109/iac.2018.8780482 · Cited by 5
Learning Data-Adaptive Interest Points through Epipolar Adaptation.
Yang, G., Malisiewicz, T., & Belongie, S. (2019) · Cited by 5
Hardware Friendly Robust Synthetic Basis Feature Descriptor
Zhang, D., Raven, L. A., Lee, D.-J., Meng, Y., & Desai, A. (2019) · Electronics · doi:10.3390/electronics8080847 · Cited by 4
“The Scale-Invariant Feature Transform (SIFT) [3] and the Speeded-Up Robust Features (SURF) [4] are arguably the two most popular and accurate intensity-based algorithms.”
Frame Stitching in Human Oral Cavity Environment Using Intraoral Camera
Ghanoum, M., Ali, A., Elshazly, S., et al. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8803071 · Cited by 5
“The KAZE algorithm showed a better performance in both detecting and describing features from normal maps against state-of-the-art methods (e.g., SIFT or SURF) .”
MethodDual Networks for High-Precision and High-Speed Registration of Brain Electron Microscopy Images
Shu, C., Tong, X., Zhou, F., Chen, X., & Han, H. (2020) · Brain Sciences · doi:10.3390/brainsci10020086 · Cited by 6
Image-matching framework based on region partitioning for target image location
Liu, X., Li, J.-B., Pan, J., Wang, S., Lv, X., & Cui, S. (2020) · Telecommunication Systems · doi:10.1007/s11235-020-00657-x · Cited by 5
Obstacle-Avoidance Algorithm Using Deep Learning Based on RGBD Images and Robot Orientation
Saleem, A. I., Jabri, K. A., Maashri, A. A., et al. (2020) · doi:10.1109/iceee49618.2020.9102526 · Cited by 5
“…extract information from the raw sensor readings based on artificially designed complex features (e.g., Scale Invariant Feature Transform (SIFT) [7, 8], Speed Up Robust Feature (SURF) [9], Binary Robust Independent Elementary Features (BRIEF) [10], and Oriented FAST and Rotated BRIEF (ORB) [11]).”
BackgroundResearch on image feature matching algorithm based on feature optical flow and corner feature
Song, T., Chen, B., Zhao, F. M., Huang, Z., & Huang, M. J. (2020) · The Journal of Engineering · doi:10.1049/joe.2019.1156 · Cited by 5
“After the scale invariant feature transform (SIFT) [9, 10] and speed-up robust feature (SURF) [11] algorithms were proposed, Rublee et al. [12] proposed a highly efficient ORB algorithm.”
MethodFast and Robust Certifiable Estimation of the Relative Pose Between Two Calibrated Cameras
García-Salguero, M., & González-Jiménez, J. (2021) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-021-01044-0 · Cited by 10
Online Video Stabilization Based on Converting Deep Dense Optical Flow to Motion Mesh
Tran, L. T., & Ly, N. Q. (2020) · NAFOSTED Conference on Information and Computer Science (NICS) · doi:10.1109/nics51282.2020.9335882 · Cited by 5
“For parametric 2D motion model (such as homography, affine) estimation method [4], first, we use methods such as SIFT [5], SURF [7], FAST [6] or Shi-Tomasi [8] method to detect features on frame , then tracking (as Pyramidal Lucas-Kanade feature tracker [9]) them on frame .”
MethodDecomposing Textures using Exponential Analysis
Hou, Y., Cuyt, A. A. M., Lee, W.-S., & Bhowmik, D. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9413909 · Cited by 8
“In support, we experiment with a texture classification task and compare with typical speeded up robust features (SURF) [21].”
MethodWang, F., Liu, Z., Zhu, H., et al. (2021) · Sensors · doi:10.3390/s21072416 · Cited by 5
Misra, I., Rohil, M. K., Moorthi, S. M., et al. (2021) · Communications in computer and information science · doi:10.1007/978-981-16-1103-2_3 · Cited by 5
Karaaslan, E., Sen, B., Ercan, T., et al. (2021) · Transportation Research Record Journal of the Transportation Research Board · doi:10.1177/03611981211026650 · Cited by 5
Multiple Master-Slave FPGA Architecture of a Stereo Visual Odometry
Chien, C.-H., Hsu, C., & Chien, C.-J. (2021) · IEEE Access · doi:10.1109/access.2021.3098856 · Cited by 3
“Features such as FAST [1] or SURF [2] are widely used.”
MethodQualitative comparison of methodologies for detecting surface defects in aircraft interiors
Mosca, N., Patruno, C., Renò, V., Nitti, M., & Stella, E. (2021) · IEEE International Workshop on Metrology for AeroSpace (MetroAeroSpace) · doi:10.1109/metroaerospace51421.2021.9511778 · Cited by 5
CapsNet meets ORB: A deformation‐tolerant baseline for recognizing distorted targets
Lin, Z., Gao, W., Jia, J., & Huang, F. (2021) · International Journal of Intelligent Systems · doi:10.1002/int.22677 · Cited by 6
“…visual descriptors (e.g., oriented fast and rotated brief [ORB], 34 scale ‐ invariant feature transform [SIFT], 35 and speeded up robust features [SURF] 36 ), VB ‐ ORB realizes feature matching on the basis of capsule feature vectors instead of commonly used pixel or DNN ‐ based feature maps.”
BackgroundHu, H., Immel, F., Janosovits, J., Lauer, M., & Stiller, C. (2021) · IEEE International Conference on Automation Science and Engineering (CASE) · doi:10.1109/case49439.2021.9551449 · Cited by 8
A Photo Identification Framework to Prevent Copyright Infringement with Manipulations
Kim, D., Heo, S., Kang, J., Kang, H.-G., & Lee, S. (2021) · Applied Sciences · doi:10.3390/app11199194 · Cited by 8
Real-time Detection of 2D Tool Landmarks with Synthetic Training Data
Vanherle, B., Put, J., Michiels, N., & Reeth, F. V. (2021) · International Conference on Robotics, Computer Vision and Intelligent Systems (ROBOVIS) · doi:10.5220/0010689900003061 · Cited by 5
GeoSOT-OctoMap: An Octree Grid Map Model for Autonomous Driving
Jian, Y., Fang, H., Pu, G., & Chen, B. (2021) · IEEE International Conference on Unmanned Systems (ICUS) · doi:10.1109/icus52573.2021.9641485 · Cited by 6
Image Splicing Detection Scheme Using Surf and Mean-LBP Based Morphological Operations
Alsandi, N. S. A. (2021) · Science Journal of University of Zakho · doi:10.25271/sjuoz.2021.9.4.866 · Cited by 5
Deeply Activated Salient Region for Instance Search
Xiao, H.-C., Zhao, W., Jie, L., Hong, Y.-G., & Ngo, C. (2022) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3510004 · Cited by 10
“Hand-crafted features such as SIFT [5] and SURF [6] that are superior in local image matching were de-facto descriptors at that time.”
MethodHigh-Precision Registration of Lunar Global Mapping Products Based on Spherical Triangular Mesh
Zheng, B., Di, K., Liu, B., et al. (2022) · Remote Sensing · doi:10.3390/rs14061442 · Cited by 5
A Tighter Relaxation for the Relative Pose Problem Between Cameras
García-Salguero, M., Briales, J., & González-Jiménez, J. (2022) · Journal of Mathematical Imaging and Vision · doi:10.1007/s10851-022-01085-z · Cited by 8
“To generate the correspondences, we extract and match 100 SURF [44] features.”
MethodStable keypoints selection for 2D LiDAR based place recognition with map data reduction
Douadi, L., Dupuis, Y., & Vasseur, P. (2022) · Robotica · doi:10.1017/s0263574722000613 · Cited by 6
Infrared and Visible Image Registration Based on Automatic Robust Algorithm
Ji, J., Zhang, Y., Lin, Z., et al. (2022) · Electronics · doi:10.3390/electronics11111674 · Cited by 5
An improved point feature‐based sparse stereo vision
Chen, C., Song, B., Bu, S., & He, L. (2022) · IET Image Processing · doi:10.1049/ipr2.12564 · Cited by 4
Zheng, Y., & Zheng, P. (2022) · doi:10.1109/icicse55337.2022.9828915 · Cited by 5
Infrared-to-Optical Image Translation for Keypoint-Based Image Registration
Elsaeidy, M., Erkol, M. E., Gunturk, B. K., et al. (2022) · Signal Processing and Communications Applications Conference (SIU) · doi:10.1109/siu55565.2022.9864815 · Cited by 5
“SIFT is one of the most-known traditional local feature descriptors in computer vision and is still widely used in many areas, along with the Speeded-Up Robust Features (SURF) [7], and Oriented FAST and Rotated BRIEF (ORB) [8].”
BackgroundJiang, W., Wu, J. T., Chen, C., Chen, J., Zeng, X., Zhong, L., Di, J., Wu, X., & Qin, Y. (2022) · Optics Express · doi:10.1364/oe.465214 · Cited by 6
Zhu, D., Zhan, W., Fu, J., Jiang, Y., Xu, X., Guo, R., & Chen, Y. (2022) · Electronics · doi:10.3390/electronics11182866 · Cited by 5
Can Post-hoc Explanations Effectively Detect Out-of-Distribution Samples?
Martínez-Seras, A., Ser, J. D., & Bringas, P. G. (2022) · IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) · doi:10.1109/fuzz-ieee55066.2022.9882726 · Cited by 10
AAM-ORB: affine attention module on ORB for conditioned feature matching
Song, S., Ai, L., Tang, P., Miao, Z., Gu, Y., & Chai, Y. (2023) · Signal Image and Video Processing · doi:10.1007/s11760-022-02452-4 · Cited by 5
Uncertainty-Driven Dense Two-View Structure From Motion
Chen, W., Kumar, S., & Yu, F. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3242153 · Cited by 7
PointFISH: Learning Point Cloud Representations for RNA Localization Patterns
Imbert, A., Mueller, F., & Walter, T. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-25069-9_32 · Cited by 7
“The computer vision community progressively replaces hand-crafted features [16,17] by deep learning features to analyze images.”
Experimental analysis of machine learning methods to detect Covid-19 from x-rays
Alkoot, F. M., Alkhedher, H. M., & Alkoot, Z. F. (2023) · Journal of Engineering Research · doi:10.1016/j.jer.2023.100063 · Cited by 7
“An effective feature extraction algorithm is the speed up robust features method [25,27].”
Hash Indexing-Based Image Matching for 3D Reconstruction
Cao, M., Jiang, H., & Zhao, H. (2023) · Applied Sciences · doi:10.3390/app13074518 · Cited by 5
Deng, Y., Deng, Y., Deng, Y., et al. (2023) · Sensors · doi:10.3390/s23073739 · Cited by 5
Unsupervised Multi-Scale-Stage Content-Aware Homography Estimation
Hou, B., Ren, J., & Yan, W. (2023) · Electronics · doi:10.3390/electronics12091976 · Cited by 5
“Most of the traditional methods for homography estimation [7,8] employ matched features, such as SIFT [9], SURF [10] and ORB [11], to establish the corresponding relationship.”
MethodGeng, X., Shi, F., Cheng, X., Jia, C., Wang, M., Chen, S., & Dai, H. (2023) · Computer Communications · doi:10.1016/j.comcom.2023.05.003 · Cited by 5
A New Ontology Convolutional Neural Network for Extorting Essential Elements in Video Mining
Ganesh, R. K., Kanthavel, R., Dhaya, R., Robinson, H., Julie, E. G., Kumar, R., Duong, P., Thong, P. H., & Sơn, L. H. (2023) · Journal of Signal Processing Systems · doi:10.1007/s11265-023-01864-w · Cited by 6
Estimation Contracts for Outlier-Robust Geometric Perception
Carlone, L. (2023) · Foundations and Trends in Robotics · doi:10.1561/2300000077 · Cited by 34
Real-Night-time Road Sign Detection by the Use of Cascade Object Detector
Shahbaz, S. J., Al-Zuky, A. A. D., & Al-Obaidi, F. E. M. (2023) · Iraqi Journal of Science · doi:10.24996/ijs.2023.64.6.43 · Cited by 5
Hu, Z., & Hu, L. (2023) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-023-01912-4 · Cited by 5
Hu, T., Fuchikami, R., & Ikenaga, T. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3295456 · Cited by 7
“Feature descriptor-based methods, such as scale-invariant feature transform (SIFT) [20], speeded-up robust features (SURF) [21], and oriented FAST and rotated BRIEF (ORB) [22], generate rotation-robust features, making them possible to be robust to rotation change.”
MethodResearch on Infrared and Visible Image Registration Algorithm for Complex Road Scenes
Wang, Y., Liang, X., & Chen, L. (2023) · IEEE Access · doi:10.1109/access.2023.3299266 · Cited by 6
Short-Long-Term Propagation-Based Video Inpainting
Li, S., Zhu, S., Huang, Y., et al. (2023) · IEEE Multimedia · doi:10.1109/mmul.2023.3309476 · Cited by 5
“After the initialization, we generate the SURF features [15] of both F t and F r .”
MethodCross-View Geo-Localization for Autonomous UAV Using Locally-Aware Transformer-Based Network
Bui, D. V., Kubo, M., & Satō, H. (2023) · IEEE Access · doi:10.1109/access.2023.3317950 · Cited by 8
Accuracy Analysis of Visual Odometer for Unmanned Rollers in Tunnels
Huang, H. M., Wang, X., Hu, Y., & Tan, P. (2023) · Electronics · doi:10.3390/electronics12204202 · Cited by 5
Self-adaptive SURF for image-to-video matching
Yang, M., Li, J., Li, Z., et al. (2023) · Signal Image and Video Processing · doi:10.1007/s11760-023-02802-w · Cited by 5
“In our test, for each frame, three candidate models are estimated by PROSAC with reprojection error 3, 5 and 8.”
Structure Similarity Preservation Learning for Asymmetric Image Retrieval
Wu, H., Wang, M., Zhou, W., & Li, H. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2023.3325728 · Cited by 9
“In addition, methods for aggregating local features [29], [30] have been explored, including Fisher vectors [31], VLAD [32], and ASMK [28] that produce global descriptors capable of scaling to large databases.”
MethodBuFF: Burst Feature Finder for Light-Constrained 3D Reconstruction
Ravendran, A., Bryson, M., & Dansereau, D. G. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3329355 · Cited by 7
Underwater visual feature matching based on attenuation invariance
Sun, J., Yu, H. L., Rao, Y., & Zhang, H. (2023) · Intelligent Marine Technology and Systems · doi:10.1007/s44295-023-00011-2 · Cited by 8
Research on Real-time Detection of Stacked Objects Based on Deep Learning
Geng, K., Qiao, J., Liu, N., Yang, Z., Zhang, R., & Li, H. (2023) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-023-02009-8 · Cited by 7
Towards Efficient Record and Replay: A Case Study in WeChat
Feng, S., Lu, H., Xiong, T., Deng, Y., & Chen, C. (2023) · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · doi:10.1145/3611643.3613880 · Cited by 10
Mirror world: creating digital twins of the space and persons from video streamings
Jiang, L. L., Cai, L., Wu, W., & Zhou, Z. (2023) · The Visual Computer · doi:10.1007/s00371-023-03193-2 · Cited by 5
“Finally, SURF [48] is used to extract feature points from background images and generate the feature database F , which will be used for subsequent real-time video registration.”
Lu, X., Yan, Y., Wei, T., & Ду, С. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00343 · Cited by 8
“For feature detection and description, there are many well-known works on handcrafted methods, such as SIFT [18], SURF [2], BRIEF [3] and ORB [26], which have good performance and are widely used in 3D computer vision tasks.”
MethodPakulev, K., Vakhitov, A., & Ferrer, G. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00878 · Cited by 10
“The most productive way to tackle this problem is modeling scale via the scale-space [20] - the backbone of some famous methods like SIFT [21], SURF [4], Harris-Laplace [23], Harris-Affine [24], KAZE [1].”
Concept-Based Anomaly Detection in Retail Stores for Automatic Correction Using Mobile Robots
Kapoor, A., Sengar, V., George, N., Vatsal, V., Gubbi, J., Balamuralidhar, P., & Pal, A. (2023) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc53992.2023.10394209 · Cited by 5
Saer, A. E., Γραμματικόπουλος, Λ., Sfikas, G., Karras, G. C., & Πέτσα, Έ. (2024) · Sensors · doi:10.3390/s24041083 · Cited by 5
“Image matching has been extensively researched, and numerous handcrafted methods exist, with the most popular ones being SIFT [10], SURF [11] and ORB [12].”
MethodUrban Visual Localization of Block-Wise Monocular Images with Google Street Views
Li, Z., Li, S., Anderson, J., & Shan, J. (2024) · Remote Sensing · doi:10.3390/rs16050801 · Cited by 6
Deep Feature Retention Module Network for Texture Classification
Park, S.-H., Ahn, S.-Y., & Lee, S.-W. (2024) · Applied Sciences · doi:10.3390/app14104011 · Cited by 11
Pre- and In-flight Performance of Terrain Relative Navigation on PIXL’s Micro Context Camera, M2020
Pedersen, D. A. K., Henneke, J., Benn, M., Denver, T., Jørgensen, P. S., Jørgensen, J. L., Wade, L. A., Denise, R. W., Elam, W. T., Beek, J. K. V., et al. (2024) · IEEE Aerospace Conference (AERO) · doi:10.1109/aero58975.2024.10521246 · Cited by 5
“Having located the features, they are described using a descriptor [25, 26, 27, 28, 29], where a balance for low computational load but maintaining good matching statistics is sought.”
MethodNgeljaratan, L., Moustafa, M. A., Sumarno, A., et al. (2024) · Infrastructures · doi:10.3390/infrastructures9060097 · Cited by 5
A High-Performance ORB Accelerator with Algorithm and Hardware Co-design for Visual Localization
Qi, X., Liu, Y., Hao, S., Liu, Z., Huang, K., Yang, M., Zhou, L., & Zhou, J. (2024) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas58744.2024.10558318 · Cited by 8
“ORB feature point[2] extraction, evolved from the SIFT[3] and SURF[4] algorithms, considerably lower computational complexity, establishing it as an efficient keypoint algorithm.”
MethodMS-NetVLAD: Multi-Scale NetVLAD for Visual Place Recognition
Uggi, A., & Channappayya, S. S. (2024) · IEEE Signal Processing Letters · doi:10.1109/lsp.2024.3425279 · Cited by 7
“Classical approaches refer back to the early stages of exploring image representation with interest points and their descriptors [13], [14], [15].”
BackgroundWang, Y., Zhang, Z., Huang, Y., & Su, Y. (2024) · The Journal of the Astronautical Sciences · doi:10.1007/s40295-024-00461-8 · Cited by 5
“7 Unity perspective with the 3D model OSIRIS-Rex, sensors and one directional light [44] with BF on the same image sequence.”
Cong, R., Wu, C., Song, X., Zhang, W., Kwong, S., Li, H., & Ji, P. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3465034 · Cited by 9
“…technology, many manual features and probabilistic graphical models are used for monocular depth estimation in the machine learning stage [25], [26], [27] And in recent years, the rise of deep learning technology has greatly facilitated the development of monocular depth estimation task, which…”
MethodMAD-DR: Map Compression for Visual Localization with Matchness Aware Descriptor Dimension Reduction
Wang, Q. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72784-9_15 · Cited by 7
Analysis of Deep Feature Matching Algorithms in UAV Visual Localization
Prutyanov, V. V., Ternov, M. A., & Kostrov, D. S. (2024) · International Russian Automation Conference (RusAutoCon) · doi:10.1109/rusautocon61949.2024.10694214 · Cited by 5
“In [24, 25], the algorithms are similar to the previous ones as the SURF method [26] was used to extract key points.”
MethodDeep Matching Online Video Stabilization Using TSNet
Wang, F., Zhong, B., Liu, T., Kong, X., Bai, H., & Wan, G. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3476524 · Cited by 9
“Compared with other keypoint detection methods such as traditional methods ORB [21], SIFT [22], and SURF [23], and deep learning-based methods L2Net [24], LF-Net [25], and RF-Net [26], SuperPoint adopts a lightweight structure and uses pixel-level positive and negative sample annotations, which can…”
MethodMulti-culture label-free quantitative cell migration sensing with single-cell precision
Arcab, P., Rogalski, M., Marzejon, M., et al. (2024) · Biomedical Optics Express · doi:10.1364/boe.541010 · Cited by 5
Context-Aware Fine-Grained Product Recognition on Grocery Shelves
Budimir, L. A., Kalafatić, Z., Subašić, M., & Lončarić, S. (2025) · IEEE Access · doi:10.1109/access.2025.3532745 · Cited by 6
“Prior to deep learning, product recognition on store shelves relied on carefully designed image processing and mathematical operations, such as SIFT [32] in [8], [22], and [33], and SURF [34] in [3].”
Sachar, S., & Kumar, A. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.111879 · Cited by 5
RGB-D camera and graph neural network-based SLAM for dynamic and low-texture environments
Liu, X., Zhang, X., Yan, X., Liu, P., & Lao, H. (2025) · Scientific Reports · doi:10.1038/s41598-025-12978-5 · Cited by 7
Optimization of Joint Detector for Ultrasound Images Using Mixtures of Image Feature Descriptors
Wereszczyński, K., Segen, J., Kulbacki, M., et al. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-15705-4_27 · Cited by 5
Scene Analysis for Service Robots
Eidenberger, R., Grundmann, T., Schneider, M., et al. (2012) · Springer tracts in advanced robotics · doi:10.1007/978-3-642-25116-0_15 · Cited by 5
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.