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 1,101–1,200 of 14,752
Bio-inspired methods modeled for respiratory disease detection from medical images
Woźniak, M., & Połap, D. (2018) · Swarm and Evolutionary Computation · doi:10.1016/j.swevo.2018.01.008 · Cited by 63
Multimodal Remote Sensing Image Registration Based on Image Transfer and Local Features
Zhang, J., Ma, W., Wu, Y., et al. (2019) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2019.2896341 · Cited by 63
An efficient image descriptor for image classification and CBIR
Shakarami, A., & Tarrah, H. (2020) · Optik · doi:10.1016/j.ijleo.2020.164833 · Cited by 67
Robust feature matching using guided local outlier factor
Wang, G., & Chen, Y. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.107986 · Cited by 65
Deep learning-based visual detection of marine organisms: A survey
Wang, N., Chen, T., Liu, S., Wang, R., Karimi, H. R., & Lin, Y. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.02.018 · Cited by 69
Wang, K., Liu, Y., & Li, L. (2014) · IEEE/ASME Transactions on Mechatronics · doi:10.1109/tmech.2014.2298247 · Cited by 63
“In our algorithm, relative displacements between feature points and the robot are adaptively estimated based on speeded up robust features (SURF) feature tracking [21] and orientation/velocity measurements from AHRS/odometry sensors, and can be theoretically proved to finally converge to their true values.”
MethodExpertise estimation based on simple multimodal features
Ochoa, X., Chiluiza, K., Méndez, G., Luzardo, G., Guamán, B., & Castells, J. (2013) · Proceedings of the 15th ACM on International conference on multimodal interaction · doi:10.1145/2522848.2533789 · Cited by 61
“An implementation of the Speeded Up Robust Features (SURF) technique [8] provided in the OpenCV library [9] was used to extract the feature points of the calculator images.”
MethodCounting crowd flow based on feature points
Liang, R., Zhu, Y., & Wang, H. (2014) · Neurocomputing · doi:10.1016/j.neucom.2013.12.040 · Cited by 63
Efficient Extraction of Robust Image Features on Mobile Devices
Chen, W.-C., Xiong, Y., Gao, J., Gelfand, N., & Grzeszczuk, R. (2007) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar.2007.4538870 · Cited by 63
“On average, we achieve approximately 30% speedup over [1], using double precision floating point numbers in both cases.”
BackgroundMethodInfluentialAn image-to-map loop closing method for monocular SLAM
Williams, B., Cummins, M., Neira, J., Newman, P. M., Reid, I. D., & Tardós, J. D. (2008) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2008.4650996 · Cited by 80
“The matching is performed by detecting in each image the presence or absence of features from a visual vocabulary [13] based on SURF features [1], which is learned off line from training data.”
MethodA hierarchical image clustering cosegmentation framework
Kim, E., Li, H., & Huang, X. (2012) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2012.6247737 · Cited by 62
“Histogram of SURF features The SURF feature [3] (Speeded Up Robust Feature) is a scale and rotation invariant detector and descriptor.”
Fast PRISM: Branch and Bound Hough Transform for Object Class Detection
Lehmann, A., Leibe, B., & Gool, L. V. (2010) · International Journal of Computer Vision · doi:10.1007/s11263-010-0342-x · Cited by 64
Self-citationRobust Visual Place Recognition with Graph Kernels
Stumm, E. S., Christopher Mei, A., Lacroix, S., Nieto, J. I., Hutter, M., & Siegwart, R. (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.2016.491 · Cited by 69
A Hole Filling Approach Based on Background Reconstruction for View Synthesis in 3D Video
Luo, G., Zhu, Y., Li, Z., & Zhang, L. (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.2016.197 · Cited by 65
“In the motion compensation process, SURF [3] is used to detect and describe the feature points in the reference frame and the current frame.”
MethodBehavior Knowledge Space-Based Fusion for Copy–Move Forgery Detection
Ferreira, A., Felipussi, S. C., Alfaro, C., Fonseca, P. A., Vargas-Muñoz, J. E., dos Santos, J. A., & Rocha, A. D. R. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2593583 · Cited by 66
Monocular localization in urban environments using road markings
Lu, Y., Huang, J., Chen, Y.-T., & Heisele, B. (2017) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2017.7995762 · Cited by 72
Multi-material classification of dry recyclables from municipal solid waste based on thermal imaging
Gundupalli, S. P., Hait, S., & Thakur, A. (2017) · Waste Management · doi:10.1016/j.wasman.2017.09.019 · Cited by 68
Data Augmentation-Based Joint Learning for Heterogeneous Face Recognition
Cao, B., Wang, N., Li, J., & Gao, X. (2018) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2018.2872675 · Cited by 67
“For SURF, we use the library function in the MATLAB R2016b software, which uses the standard SURF-64 version.”
Kong, X., & Li, J. (2018) · Automation in Construction · doi:10.1016/j.autcon.2018.12.011 · Cited by 66
Huang, H., & Ciou, A.-J. (2019) · EURASIP Journal on Image and Video Processing · doi:10.1186/s13640-019-0469-9 · Cited by 67
Write-a-video: computational video montage from themed text
Wang, M., Yang, G., Hu, S., Yau, S., & Shamir, A. (2019) · ACM Transactions on Graphics · doi:10.1145/3355089.3356520 · Cited by 92
“A shot boundary is marked between frames if their histograms in HSV color space differ by more than 80%, and more than 80% of tracked SURF keypoints [Bay et al. 2006] fail to match between the frames.”
BackgroundMethodRealistic Modeling of Water Droplets for Monocular Adherent Raindrop Recognition Using Bézier Curves
Roser, M., Kurz, J., & Geiger, A. (2011) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-22819-3_24 · Cited by 63
“It can be achieved densely as wells as from preselected points of interest, like CenSurE [1] or SURF [2].”
BackgroundInfluentialAugmented reality applications for cultural heritage using Kinect
Bostanci, E., Kanwal, N., & Clark, A. F. (2015) · Human-centric Computing and Information Sciences · doi:10.1186/s13673-015-0040-3 · Cited by 62
Photo Album Compression for Cloud Storage Using Local Features
Shi, Z., Sun, X., & Wu, F. (2014) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2014.2298291 · Cited by 61
“Speed up robust feature (SURF) descriptors reduce the dimensions of the feature vector to 64 with a performance similar to SIFT [21].”
MethodM anta M atcher: automated photographic identification of manta rays using keypoint features
Town, C. P., Marshall, A. D., & Sethasathien, N. (2013) · Ecology and Evolution · doi:10.1002/ece3.587 · Cited by 78
“Features are extracted using parameters as described in the literature (Bay et al. 2006; Rublee et al. 2011) and compared using “brute-force” matching (i.e., not using the BBF speedup used for the MM and SIFT methods) with the distance criterion being the Euclidean norm.”
BackgroundMethodInfluentialModeling Natural Images Using Gated MRFs
Ranzato, M., Mnih, V., Susskind, J. M., & Hinton, G. E. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2013.29 · Cited by 61
“Most methods [2] start by applying some well-engineered features, like SIFT [3], HoG [4], SURF [5], or PHoG [6], to describe image patches, and then aggregate these features at different spatial resolutions and on different parts of the image to produce a feature vector, which is subsequently fed…”
3D Extended Histogram of Oriented Gradients (3DHOG) for Classification of Road Users in Urban Scenes
Buch, N., Orwell, J., & Velastín, S. A. (2009) · British Machine Vision Conference (BMVC) · doi:10.5244/c.23.15 · Cited by 62
Estimation and Detection with Applications to Navigation
Törnqvist, D. (2008) · DiVA (Linkoping University) · Cited by 66
Keep it brief: Scalable creation of compressed localization maps
Dymczyk, M., Lynen, S., Bosse, M. C., & Siegwart, R. (2015) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2015.7353722 · Cited by 57
“Given a novel image together with features [2, 3] extracted from it, the camera pose w.”
BackgroundObject Detection and Recognition for Assistive Robots: Experimentation and Implementation
Martínez-Martín, E., & del Pobil, A. P. (2017) · IEEE Robotics & Automation Magazine · doi:10.1109/mra.2016.2615329 · Cited by 72
Copy-move forgery detection using combined features and transitive matching
Lin, C., Lu, W., Huang, X., Liu, K., Sun, W., Lin, H., & Tan, Z. (2018) · Multimedia Tools and Applications · doi:10.1007/s11042-018-6922-4 · Cited by 67
“Feature exaction methods such as the Scale-Invariant Feature Transform (SIFT) [45] and the Speeded Up Robust Features (SURF) [6] are most widely used in keypoint-based schemes.”
MethodA Deep Cross-Modality Hashing Network for SAR and Optical Remote Sensing Images Retrieval
Xiong, W., Xiong, Z., Zhang, Y., Cui, Y., & Gu, X. (2020) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2020.3021390 · Cited by 65
“Most existing feature extractors are based on low-level visual features, including global features related to spectral (color) [17], texture [19], shape [22], as well as local features based on scale invariant feature transform (SIFT) [23], difference of Gaussian (DoG) [24], and speeded up robust features (SURF) [25].”
BackgroundUsing Deep Learning for Defect Classification on a Small Weld X-ray Image Dataset
Ajmi, C., Zapata, J., Martínez-Álvarez, J. J., Doménech‐Asensi, G., & Ruíz, R. (2020) · Journal of Nondestructive Evaluation · doi:10.1007/s10921-020-00719-9 · Cited by 73
“…histograms, color correlograms [12]), keypoint detectors (FAST [27], Harris [9], DoG [19], etc) and local invariant descriptors (SIFT [19], SURF [1], BRIEF [2], ORB [4], to name a few too), but now we can use raw pixel intensities as inputs to our machine learning models, as is now common with…”
MethodHigh Efficiency Compression for Object Detection
Choi, H., & Bajić, I. V. (2018) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2018.8462653 · Cited by 61
“Another recent work [6] develops a rate control scheme for H.264/AVC video coding that preserves SIFT [7] and SURF [8] features, which have also been found useful in many computer vision problems.”
MethodAdaptive neuro-heuristic hybrid model for fruit peel defects detection
Woźniak, M., & Połap, D. (2017) · Neural Networks · doi:10.1016/j.neunet.2017.10.009 · Cited by 61
Secure Image LBP Feature Extraction in Cloud-Based Smart Campus
Xia, Z., Ma, X., Shen, Z., Sun, X., Xiong, N. N., & Jeon, B. (2018) · IEEE Access · doi:10.1109/access.2018.2845456 · Cited by 63
“Considering the similarity between HOG (histogram of oriented gradients), SURF (speeded up robust features) and SIFT features in convolution operations and comparison operations, in their other works [22], [23], the authors applied these two protocols to implement the secure extraction of HOG [24] and SURF [25].”
MethodGlobal Localization with Object-Level Semantics and Topology
Liu, Y., Pétillot, Y., Lane, D. M., & Wang, S. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794475 · Cited by 79
StructDepth: Leveraging the structural regularities for self-supervised indoor depth estimation
Li, B., Huang, Y., Liu, Z., Zou, D., & Yu, W. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01243 · Cited by 86
“[51] adopted an optical-flow-based training paradigm supervised by the flow field from an optical flow network, initialized from sparse SURF [1] correspondences.”
InfluentialMaurer, J., & Rupper, S. (2015) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2015.06.008 · Cited by 67
RGB-D Object Recognition via Incorporating Latent Data Structure and Prior Knowledge
Tang, J., Jin, L., Li, Z., & Gao, S. (2015) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2015.2476660 · Cited by 60
“However, these local features, such as SIFT [30], HOG [10] and SURF [1], are invariant to the common variance.”
MethodMoving Objects Detection and Tracking Framework for UAV-based Surveillance
Ibrahim, A. P. O. H., Pang, W. C., Seet, G. L., Lau, W. S. M., & Czajewski, W. (2010) · Pacific-Rim Symposium on Image and Video Technology · doi:10.1109/psivt.2010.83 · Cited by 63
Ishiguro, Y., Mujibiya, A., Miyaki, T., et al. (2010) · Augmented Human International Conference · doi:10.1145/1785455.1785480 · Cited by 61
“We use fast object recognition by using the SURF [1] descriptor for matching images that is limited to the gazed area with the past-images database (Figure 5).”
MethodMobile Visual Search: Architectures, Technologies, and the Emerging MPEG Standard
Girod, B., Chandrasekhar, V., Grzeszczuk, R., & Reznik, Y. A. (2011) · IEEE Multimedia · doi:10.1109/mmul.2011.48 · Cited by 68
Learning Discriminative Correlation Subspace for Heterogeneous Domain Adaptation
Yan, Y., Li, W., Ng, M. K., Tan, M., Wu, H., Min, H., & Wu, Q. (2017) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2017/454 · Cited by 58
Distributed inference-based multi-robot exploration
Smith, A. J., & Hollinger, G. A. (2018) · Autonomous Robots · doi:10.1007/s10514-018-9708-7 · Cited by 70
“not used (such as SURF [17] or SIFT [18] features) to identify map structure because visual features in maps are often highly similar and lead to ambiguity in matching.”
MethodVision-Based Pose Estimation for Textureless Space Objects by Contour Points Matching
Zhang, X., Jiang, Z., Zhang, H., & Wei, Q. (2018) · IEEE Transactions on Aerospace and Electronic Systems · doi:10.1109/taes.2018.2815879 · Cited by 65
Composite components-based face sketch recognition
Liu, D., Li, J., Wang, N., et al. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.03.042 · Cited by 61
Brunetti, A., Carnimeo, L., Trotta, G. F., & Bevilacqua, V. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.06.080 · Cited by 63
Joint Embedding Learning and Low-Rank Approximation: A Framework for Incomplete Multiview Learning
Tao, H., Hou, C., Yi, D., Zhu, J., & Hu, D. (2019) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2019.2953564 · Cited by 65
“SIFT (200), SURF (200) [41], and LBP (256) features are extracted.”
BackgroundMonocular Visual Odometry Based on Depth and Optical Flow Using Deep Learning
Ban, X., Wang, H., Chen, T., Wang, Y., & Xiao, Y. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2020.3024011 · Cited by 67
“In computer vision, the frequently used feature detection methods are FAST [17], SURF [18], ORB [19], and BRIEF [20].”
MethodProduction Rules for Stereo Acquisition
Zilly, F., Kluger, J., & Kauff, P. P. (2011) · Proceedings of the IEEE · doi:10.1109/jproc.2010.2095810 · Cited by 60
“Any suitable feature detectors like SIFT or SURF can be used for this purpose [38], [39].”
MethodFeature Representations for Scene Text Character Recognition: A Comparative Study
Yi, C., Yang, X., & Tian, Y. (2013) · IAPR International Conference on Document Analysis and Recognition (ICDAR) · doi:10.1109/icdar.2013.185 · Cited by 70
Coral reef image classification employing Improved LDP for feature extraction
Mary, N. A. B., & Dharma, D. (2017) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2017.09.008 · Cited by 60
Fast bridge deflection monitoring through an improved feature tracing algorithm
Yu, S., & Zhang, J. (2019) · Computer-Aided Civil and Infrastructure Engineering · doi:10.1111/mice.12499 · Cited by 61
CGA: a new feature selection model for visual human action recognition
Guha, R., Khan, A. H., Singh, P. K., Sarkar, R., & Bhattacharjee, D. (2020) · Neural Computing and Applications · doi:10.1007/s00521-020-05297-5 · Cited by 66
“As mentioned earlier, a set of four different feature descriptors, namely HOG [24], GLCM [25], SURF [26], and GIST [27], are extracted from the video frames in the present work.”
BackgroundMethodA Review of Computer Vision Methods in Network Security
Zhao, J., Masood, R., & Seneviratne, S. (2021) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2021.3086475 · Cited by 81
LocalTrans: A Multiscale Local Transformer Network for Cross-Resolution Homography Estimation
Shao, R., Wu, G., Zhou, Y., Fu, Y., Fang, L., & Liu, Y. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01462 · Cited by 78
RelPose: Predicting Probabilistic Relative Rotation for Single Objects in the Wild
Zhang, J. Y., Ramanan, D., & Tulsiani, S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19821-2_34 · Cited by 119
A Computer Vision Technique to Detect Phishing Attacks
Rao, R. S., & Ali, S. T. (2015) · International Conference on Communication Systems and Network Technologies · doi:10.1109/csnt.2015.68 · Cited by 73
Artificial landmark-based underwater localization for AUVs using weighted template matching
Kim, D., Lee, D., Myung, H., & Choi, H. (2014) · Intelligent Service Robotics · doi:10.1007/s11370-014-0153-y · Cited by 60
“Various techniques, including SIFT [32], SURF [33], GLOH [34], and HOG [35], are available to analyze the correspondences between the feature points detected in two different images.”
MethodPerspective-SIFT: An efficient tool for low-altitude remote sensing image registration
Cai, G., Jodoin, P., Li, S., Wu, Y., Su, S., & Huang, Z. (2013) · Signal Processing · doi:10.1016/j.sigpro.2013.04.008 · Cited by 60
Towards Wiki-based Dense City Modeling
Irschara, A., Zach, C., & Bischof, H. (2007) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2007.4409216 · Cited by 61
“We adopt the idea of [2] and match features with the same contrast only, by taking advantage of the Laplacian sign.”
MethodSünderhauf, N., Lange, S., & Protzel, P. (2007) · IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR) · doi:10.1109/ssrr.2007.4381265 · Cited by 76
“Landmark detection and matching can be performed with sophisticated and robust features like SURF [21] (we successfully applied SURF features in a FastSLAM context very recently [22] ) or with a more “quick and dirty”-like solution with the Harris detector [23] paired with a correlation based matching.”
MethodDASC: Robust Dense Descriptor for Multi-Modal and Multi-Spectral Correspondence Estimation
Kim, S., Min, D., Ham, B., Do, M. N., & Sohn, K. (2016) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2615619 · Cited by 63
Yuan, X., Martínez-Ortega, J.-F., Eckert, M., & López, L. (2016) · Sensors · doi:10.3390/s16071148 · Cited by 63
Computer Vision based Assistive Technology for Blind and Visually Impaired People
Sivan, S., & Darsan, G. (2016) · International Conference on Computing, Communication and Networking Technologies (ICCCNT) · doi:10.1145/2967878.2967923 · Cited by 71
“BRISK was preferred over SURF [13] and SIFT [16] due to its high performance, also offers a much easily configurable circular sampling pattern which computes th brightness comparisons to form a binary descriptor string.”
MethodUnderwater cave mapping using stereo vision
Weidner, N., Rahman, S., Li, A. Q., & Rekleitis, I. M. (2017) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2017.7989672 · Cited by 78
“For every point on the contour of the left image a SURF feature descriptor [39] is calculated using the left rectified image.”
BackgroundMethodInfluentialVision-Based Target Three-Dimensional Geolocation Using Unmanned Aerial Vehicles
Zhang, L., Deng, F. Y., Chen, J., Bi, Y., Phang, S. K., Chen, X., & Chen, B. M. (2018) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2018.2807401 · Cited by 62
RPNet: An End-to-End Network for Relative Camera Pose Estimation
En, S., Lechervy, A., & Jurie, F. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-11009-3_46 · Cited by 66
On-the-Fly Olive Tree Counting Using a UAS and Cloud Services
Salamí, E., Gallardo, A., Skorobogatov, G., & Barrado, C. (2019) · Remote Sensing · doi:10.3390/rs11030316 · Cited by 65
Indian Traffic Sign Detection and Recognition
Alam, A., & Jaffery, Z. A. (2019) · International Journal of Intelligent Transportation Systems Research · doi:10.1007/s13177-019-00178-1 · Cited by 60
RGB-D-Based Object Recognition Using Multimodal Convolutional Neural Networks: A Survey
Gao, M., Jiang, J., Zou, G., John, V., & Liu, Z. (2019) · IEEE Access · doi:10.1109/access.2019.2907071 · Cited by 60
Corner Detection Using Multi-directional Structure Tensor with Multiple Scales
Zhang, W., & Sun, C. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01257-2 · Cited by 62
“This is impossible for the existing corner detectors (Deriche and Giraudon 1993; Smith and Brady 1997; Rosten et al. 2010; Shui and Zhang 2013; Xia et al. 2014; Moravec 1979; Harris and Stephens 1988; Noble 1988; Gårding and Lindeberg 1996; Lindeberg 1998; Schmid et al. 2000; Kenney et al. 2003; Mikolajczyk and Schmid…”
BackgroundMethodResultInfluentialAn autonomous robotic platform for automatic extraction of detailed semantic models of buildings
Adán, A., Quintana, B., Prieto, S. A., & Bosché, F. (2019) · Automation in Construction · doi:10.1016/j.autcon.2019.102963 · Cited by 64
“We conducted a set of experiments using SIFT ([67]) and SURF ([68]) algorithms using a database composed of 31 different models of BSCs.”
MethodEnd-to-end multimodal image registration via reinforcement learning
Hu, J., Luo, Z., Wang, X., Sun, S., Yin, Y., Cao, K., Song, Q., Lyu, S., & Wu, X. (2020) · Medical Image Analysis · doi:10.1016/j.media.2020.101878 · Cited by 76
Dayoub, F., Cielniak, G., & Duckett, T. (2011) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2011.02.013 · Cited by 59
“Local features including SIFT [14], SURF [15], MSER [16], etc.”
BackgroundMethodWang, S., & Li, W. (2021) · Computers, Environment and Urban Systems · doi:10.1016/j.compenvurbsys.2021.101715 · Cited by 59
Zhao, K., Jin, X., Ji, J., Wang, J., Ma, H., & Zhu, X. (2019) · Biosystems Engineering · doi:10.1016/j.biosystemseng.2019.03.004 · Cited by 67
Real-Time Dense Reconstruction of Tissue Surface From Stereo Optical Video
Zhou, H., & Jayender, J. (2019) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2019.2927436 · Cited by 79
A multiple camera position approach for accurate displacement measurement using computer vision
Kromanis, R., & Kripakaran, P. (2021) · Journal of Civil Structural Health Monitoring · doi:10.1007/s13349-021-00473-0 · Cited by 59
Convolutional Hough Matching Networks
Min, J., & Cho, M. (2021) · 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/cvpr46437.2021.00296 · Cited by 61
Pedestrian and Vehicle Detection in Autonomous Vehicle Perception Systems—A Review
Galvao, L. G., Abbod, M., Kalganova, T., Palade, V., & Huda, M. N. (2021) · Sensors · doi:10.3390/s21217267 · Cited by 59
A Sim-to-Real Object Recognition and Localization Framework for Industrial Robotic Bin Picking
Li, X., Cao, R., Feng, Y., Chen, K., Yang, B., Fu, C., Li, Y., Dou, Q., Liu, Y., & Heng, P. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3149026 · Cited by 74
“Alternatively, sparse-feature-based methods [10], [27], [30] first extract points of interest from the input image and then compute local descriptors (e.g., SIFT [28] and SURF [3]) for database matching.”
A review of deep learning methods for digitisation of complex documents and engineering diagrams
Jamieson, L., Moreno‐García, C. F., & Elyan, E. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10779-2 · Cited by 91
“Examples of commonly used image features include Histogram of Oriented Gradient (HOG) (Dalal and Triggs 2005), Scale Invariant Feature Transform (SIFT) (Lowe 2004), Speeded Up Robust Features (SURF) (Bay et al. 2006) and Local Binary Pattern (LBP) (Ojala et al. 2002).”
Vehicle Ego-Localization by Matching In-Vehicle Camera Images to an Aerial Image
Noda, M., Takahashi, T., Deguchi, D., et al. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-22819-3_17 · Cited by 59
Vehicle Detection from Aerial Images Using Local Shape Information
Choi, J.-Y., & Yang, Y.-K. (2009) · Lecture notes in computer science · doi:10.1007/978-3-540-92957-4_20 · Cited by 63
“Invariant features are robust to image scale, rotation, and partially invariant to changing viewpoints, and change in illumination [3].”
BackgroundLoop closure detection for visual SLAM systems using deep neural networks
Gao, X., & Zhang, T. (2015) · Chinese Control Conference (CCC) · doi:10.1109/chicc.2015.7260555 · Cited by 59
The SmartVision Navigation Prototype for Blind Users
-, J. M. H. D. B., -, J. B., -, J. M. F. R., -, H. P., -, M. F., -, H. F., -, J. J., -, V. T., & -, M. S. (2011) · International Journal of Digital Content Technology and its Applications · doi:10.4156/jdcta.vol5.issue5.39 · Cited by 59
“To this purpose we use the well-known algorithm coined SURF (Speeded Up Robust Features), which is a scale- and rotation-invariant interest-point detector and descriptor [3].”
MethodWon, D. H., Lee, E., Heo, M.-B., Lee, S., Lee, J., Kim, J., Sung, S., & Lee, Y. J. (2014) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2014.2304365 · Cited by 62
“It is assumed that the feature point is fixed on the navigation frame, and can be extracted and tracked by image processing methods such as Kanade–Lucas–Tomasi (KLT) [21], scale-invariant feature transform [22], and speeded up robust feature [23].”
MethodDémesticha, S., Skarlatos, D. P., & Neophytou, A. (2014) · Journal of Field Archaeology · doi:10.1179/0093469014z.00000000077 · Cited by 59
Global Gaussian approach for scene categorization using information geometry
Nakayama, H., Harada, T., & Kuniyoshi, Y. (2010) · 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.2010.5539921 · Cited by 57
“In addition, the result shows that SURF is superior in LSP15 and SIFT is superior in 8-sports.”
MethodInfluentialHand gesture recognition using Bag-of-features and multi-class Support Vector Machine
Dardas, N. H., Chen, Q., Georganas, N. D., & Petriu, E. M. (2010) · IEEE International Workshop on Haptic Audio visual Environments and Games · doi:10.1109/have.2010.5623982 · Cited by 59
“Keypoints are detected by robust feature detection methods like SIFT [2], its variant Principal Component Analysis (PCA)-SIFT [14] and SURF (Speeded Up Robust Features) [15].”
MethodA Novel Blending Technique for Underwater Gigamosaicing
Prados, R., García, R., Gracias, N., Escartin, J., & Neumann, L. (2012) · IEEE Journal of Oceanic Engineering · doi:10.1109/joe.2012.2204152 · Cited by 67
Video stabilization using Speeded Up Robust Features
Pinto, B., & Purushothaman, A. (2011) · International Conference on Communications and Signal Processing (ICCSP) · doi:10.1109/iccsp.2011.5739378 · Cited by 59
A One-Year Introductory Robotics Curriculum for Computer Science Upperclassmen
Correll, N., Wing, R. M., & Coleman, D. (2012) · IEEE Transactions on Education · doi:10.1109/te.2012.2220774 · Cited by 58
“Students are exposed to state-of-the-art search and path-finding algorithms such as Rapidly-exploring Random Trees (RRT), computationally more efficient solutions to the SLAM problem (Graph-based SLAM), advanced feature recognition algorithms such as SIFT and SURF, and feature recognition in 3-”
MethodRadar SLAM using visual features
Callmer, J., Törnqvist, D., Gustafsson, F., Svensson, H., & Carlbom, P. (2011) · EURASIP Journal on Advances in Signal Processing · doi:10.1186/1687-6180-2011-71 · Cited by 67
A Geometric Particle Filter for Template-Based Visual Tracking
Kwon, J., Lee, H. S., Park, F. C., & Lee, K. M. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2013.170 · Cited by 58
“28% ), FERNS ( 44 .”
MethodInfluentialGuided Image Contrast Enhancement Based on Retrieved Images in Cloud
Wang, S., Gu, K., Ma, S., et al. (2015) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2015.2510326 · Cited by 59
“Locating keypoints and extracting descriptors can be an alternative solution, such as computing the well-known Scale-invariant feature transform (SIFT) [35] and Speeded Up Robust Features (SURF) [36] descriptors.”
MethodLowry, S., & Milford, M. J. (2016) · IEEE Transactions on Robotics · doi:10.1109/tro.2016.2545711 · Cited by 59
Third Eye: A Shopping Assistant for the Visually Impaired
Zientara, P. A., Lee, S., Smith, G. H., Brenner, R., Itti, L., Rosson, M. B., Carroll, J. M., Irick, K. M., & Narayanan, V. (2017) · Computer · doi:10.1109/mc.2017.36 · Cited by 59
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.