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 6,401–6,500 of 14,752
Twisted window search for efficient shape localization
Gu, S., Zheng, Y., & Tomasi, C. (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.2012.6247672 · Cited by 7
“More sophisticated, well engineered, high dimensional features such as SIFT [23] or SURF [5] should enhance the robustness of feature matching in object tracking, but with a greater running time.”
Coupling Spatiotemporal Disease Modeling with Diagnosis
Mubangizi, M., Ikae, C., Spiliopoulou, A., & Quinn, J. A. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v26i1.8180 · Cited by 7
A Source Analysis of the Conficker Outbreak from a Network Telescope
Irwin, B. V. W. (2013) · SAIEE Africa Research Journal · doi:10.23919/saiee.2013.8531865 · Cited by 7
Finding Correspondences Between Images using Descriptors and Graphs
Zakharov, A., Tuzhilkin, A., & Львович, Ж. А. (2015) · Procedia Engineering · doi:10.1016/j.proeng.2015.12.131 · Cited by 9
Image-based localization for an indoor VR/AR construction training system
Carozza, L., Bosché, F., & Abdel‐Wahab, M. (2013) · Cited by 10
Local N-ary Patterns: a local multi-modal descriptor for place categorization
Jung, H., Mozos, Ó. M., Iwashita, Y., & Kurazume, R. (2016) · Advanced Robotics · doi:10.1080/01691864.2015.1120242 · Cited by 7
“A scene image is recognized as a semantic label of an image, and the visual approach of place categorization is conducted using a camera and various feature descriptors, such as GIST [5], SIFT [10,11], SURF [12], and CENTRIST [13].”
MethodObservation support system of an ROV for underwater archaeology
Sakagami, N., Takemura, F., Ono, R., et al. (2015) · International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS) · doi:10.1109/iciibms.2015.7439498 · Cited by 7
Target-Free Automatic Point Clouds Registration Using 2D Images
Gai, M., Cho, Y. K., & Xu, Q. (2013) · ASCE International Conference on Computing in Civil Engineering (i3CE) · doi:10.1061/9780784413029.108 · Cited by 7
“Encourage by Scale-invariant feature transform (SIFT) algorithm, Bay et al. (2006) presented Speeded-Up Robust Features (SURF) method using detection, description, matching scheme.”
MethodVehicle video detection and tracking quality analysis
Кустикова, В. Д., & Gergel, V. (2016) · Pattern Recognition and Image Analysis · doi:10.1134/s1054661816010156 · Cited by 7
“The operation is intended for building sets of feature points and their SURF descriptors [14] in every image, and for complete matching of the descriptors with the following cutoff of outliers by RANSAC [15].”
Local Visual Feature Detection and Description for Non-Rigid 3D Objects
Zeng, K., Wu, N., Wang, L., & Yen, K. K. (2016) · Advances In Image and Video Processing · doi:10.14738/aivp.42.1960 · Cited by 7
“SURF, designed by Bay et al. [12], is similar to SIFT with faster feature detection and description.”
MethodORB in 5 ms: An Efficient SIMD Friendly Implementation
Viswanath, P., Swami, P., Desappan, K., et al. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16628-5_48 · Cited by 7
“The most commonly used methods are Scale Invariant Feature Transform(SIFT) [1], Speeded up Robust Features(SURF) [2] and Oriented FAST and Rotated BRIEF(ORB) [3].”
MethodReiss, M. L. L., da Rocha, R. S., Ferraz, R. S., et al. (2016) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xli-b5-315-2016 · Cited by 7
“Another widely used algorithm for matching pictures is called SURF (Speeded-Up Robust Features) (Bay et al, 2008; Barazzetti et al, 2010), which is a variant of SIFT (Bay et al, 2008).”
MethodStereo reconstruction using top-down cues
Hadfield, S., Lebeda, K., & Bowden, R. (2016) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2016.08.001 · Cited by 8
“The most prevalent of these approaches is the matching of SIFT or similar features [9, 10].”
Search Space Reduction in Biometric Databases
Kavati, I., Prasad, M. V. N. K., & Bhagvati, C. (2016) · Advances in information security, privacy, and ethics book series · doi:10.4018/978-1-5225-0703-1.ch011 · Cited by 7
Lavania, C., Thulasidasan, S., LaMarca, A., Scofield, J., & Bilmes, J. (2016) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/2971648.2971716 · Cited by 8
“There is no dearth of feature extractors for images in computer vision literature, including SIFT [35], SURF [3], HOG [13] etc.”
MethodThe CUDA LATCH Binary Descriptor: Because Sometimes Faster Means Better
Parker, C., Daiter, M., Omar, K., Levi, G., & Hassner, T. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-49409-8_57 · Cited by 9
Vision Based Displacement Detection for Stabilized UAV Control on Cloud Server
Jeong, H.-J., Choi, J. D., & Ha, Y.-G. (2016) · Mobile Information Systems · doi:10.1155/2016/8937176 · Cited by 7
“This concept is expressed as a time series [23]:”
BackgroundClassifying Bacterial Species using Computer Vision and Machine Learning
Vijaykumar, V. (2016) · International Journal of Computer Applications · doi:10.5120/ijca2016911851 · Cited by 7
Abdelsamea, M. M. (2016) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2016.10.011 · Cited by 7
An improved ICP algorithm for kinect point cloud registration
Ye, Q., Yao, Y., Gui, P., et al. (2016) · International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) · doi:10.1109/fskd.2016.7603507 · Cited by 7
“In close–range image matching, SIFT in [13] and SURF in [14] are popular matching algorithms with scale–invariant and rotation– invariant feature.”
BackgroundSIFT and SURF feature analysis in visible and infrared imaging for UAVs
Li, X., & Aouf, N. (2012) · IEEE International Conference on Cybernetic Intelligent Systems (CIS) · doi:10.1109/cis.2013.6782158 · Cited by 7
Automatic image annotation using SURF descriptors
Anees, V. M., Kumar, G. S., & Sreeraj, M. (2012) · IEEE India Conference · doi:10.1109/indcon.2012.6420748 · Cited by 7
Image-based localization using Gaussian processes
López-Antequera, M., Petkov, N., & González-Jiménez, J. (2016) · International Conference on Indoor Positioning and Indoor Navigation · doi:10.1109/ipin.2016.7743697 · Cited by 7
“Most of these global description models use visual bagsof-words to describe images by collecting histograms of local features such as SURF [2].”
MethodStereo-Image Normalization of Voluminous Objects Improves Textile Defect Recognition
Siegmund, D., Kuijper, A., & Braun, A. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-50835-1_17 · Cited by 8
Srivastava, S., & Lall, B. (2016) · Indian Conference on Computer Vision, Graphics and Image Processing · doi:10.1145/3009977.3010009 · Cited by 12
Hybrid Tracking and Matching Algorithm for Mosaicking Multiple Surgical Views
Takada, C., Suzuki, T., Afifi, A., et al. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-54057-3_3 · Cited by 7
Training a New Instrument to Measure Cotton Fiber Maturity Using Transfer Learning
Turner, C., Sari‐Sarraf, H., & HEQUET, E. F. (2017) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2017.2666203 · Cited by 8
“, speeded up robust features [18] in one domain and scale- invariant feature transform [19] in the other) and report between 25–53% classification”
BackgroundEfficient segmentation for Region-based Image Retrieval using Edge Integrated Minimum Spanning Tree
Liu, Y., Huang, L., Wang, S., et al. (2016) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2016.7899918 · Cited by 7
“Intuitively, RBIR works in a scale half-way between global features like Color and Edge Directivity Descriptor (CEDD) [1] and key-point-based local features such as SIFT [2] and SURF [3].”
BackgroundIris Recognition using feature optimization
Charan, S. G. (2016) · International Conference on Applied and Theoretical Computing and Communication Technology · doi:10.1109/icatcct.2016.7912095 · Cited by 7
“Once the circle is predominantly highlighted using the template matching and blob matching (blob matching was performed with Speeded Up Robust Features (SURF) [5], [6] and cross-correlation coefficient) we can extract out the central-exterior region of eye ”iris”.”
MethodInfluentialAnalysis of Feature Point Distributions for Fast Image Mosaicking Algorithms
Behrens, A., & Röllinger, H. (2010) · Acta Polytechnica · doi:10.14311/1219 · Cited by 7
“To assure constant computational complexity, only these features are then characterized by the 64-dimensional SURF descriptor and passed to the matching process.”
MethodInfluentialEXTENDED BAG-OF-WORDS FORMALISM FOR IMAGE CLASSIFICATION
Avila, S., Thome, N., Cord, M., Junior, E. A. D. V., & Araújo, A. D. A. (2013) · HAL · Cited by 7
Solving Avatar Captchas Automatically
Korayem, M., Mohamed, A. A., Crandall, D., et al. (2012) · Communications in computer and information science · doi:10.1007/978-3-642-35326-0_11 · Cited by 7
Automated Assembly Using 3D and 2D Cameras
Kleppe, A. L., Bjørkedal, A., Larsen, K. S., & Egeland, O. (2017) · Robotics · doi:10.3390/robotics6030014 · Cited by 6
“The stability was first tested using SIFT, and it was then compared to using a hybrid algorithm, where SIFT keypoints were used with the Speeded Up Robust Feature (SURF) descriptor [18].”
MethodHard exudates referral system in eye fundus utilizing speeded up robust features
Naqvi, S. A. G., Zafar, H. M. F., & Haq, I. U. (2017) · International Journal of Ophthalmology · doi:10.18240/ijo.2017.07.24 · Cited by 7
Object State Recognition for Automatic AR-Based Maintenance Guidance
Dvorak, P., Jošth, R., & Delponte, E. (2017) · IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2017.164 · Cited by 7
Viquez, K. D. C., Arandjelović, O., Blaikie, A. J., et al. (2017) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2017.19 · Cited by 7
Underwater image matching by incorporating structural constraints
Yang, X., Liu, Z.-Y., Qiao, H., Song, Y., Ren, S., Ji, D., & Zheng, S. (2017) · International Journal of Advanced Robotic Systems · doi:10.1177/1729881417738100 · Cited by 7
“The main drawback of SIFT is its high computational expense, and quite a number of methods try to improve it by simplifying its complex extractor and descriptor, such as the principal component analysis-SIFT,(12) gradient location orientation histogram(13) and the well-known speeded up robust feature (SURF).(14) Parti…”
MethodObject Detection Technique for Small Unmanned Aerial Vehicle
Ramli, M. F., Legowo, A., & Shamsudin, S. S. (2017) · IOP Conference Series Materials Science and Engineering · doi:10.1088/1757-899x/260/1/012040 · Cited by 7
A fusion algorithm of visual odometry based on feature-based method and direct method
Feng, J., Zhang, C., Sun, B., & Song, Y. H. (2017) · Chinese Automation Congress · doi:10.1109/cac.2017.8243070 · Cited by 6
“Common features are Harris corner, SIFT [3], SURF [4], ORB [5] and so on.”
A Motion-Driven Approach for Fine-Grained Temporal Segmentation of User-Generated Videos
Apostolidis, K., Apostolidis, E., & Mezaris, V. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-73603-7_3 · Cited by 7
“[9] studies several approaches for optical flow field calculation, that include the matching of local descriptors (i.e. SIFT [21], SURF [4]) based on a variety of block matching algorithms, and the use of the PLK algorithm [6].”
MethodInfluentialAutomatic object searching by a mobile robot with single RGB-D camera
Shim, V. A., Miaolong, Y., & Tan, B. H. (2017) · Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) · doi:10.1109/apsipa.2017.8282002 · Cited by 8
Keypoint Descriptors in SIFT and SURF for Face Feature Extractions
Pui, S., & Minoi, J.-L. (2018) · Lecture notes in electrical engineering · doi:10.1007/978-981-10-8276-4_7 · Cited by 7
Local Patch Vectors Encoded by Fisher Vectors for Image Classification
Chen, S., Liu, H., Zeng, X., Qian, S., Wei, W., Wu, G., & Duan, B. (2018) · Information · doi:10.3390/info9020038 · Cited by 7
Amor, M. B., Kammoun, F., & Masmoudi, N. (2018) · Computer Standards & Interfaces · doi:10.1016/j.csi.2018.04.003 · Cited by 7
Temporal Aggregation of Visual Features for Large-Scale Image-to-Video Retrieval
García, N. (2018) · International Conference on Multimedia Retrieval · doi:10.1145/3206025.3206083 · Cited by 8
“For evaluation, as long as the retrieved frame shares strong similarities with the ground truth frame by matching SURF [7] features, it is consider a Visual Match .”
MethodImproving texture extraction and classification using smoothed morphological operators
Neiva, M. B., Vacavant, A., & Bruno, O. M. (2018) · Digital Signal Processing · doi:10.1016/j.dsp.2018.06.001 · Cited by 7
The Classification of Traditional Chinese Painting Based on CNN
Meng, Q., Zhang, H., Zhou, M., et al. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-00009-7_22 · Cited by 7
Feature Matching Improvement through Merging Features for Remote Sensing Imagery
Karim, S., Zhang, Y., Brohi, A. A., & Asif, M. R. (2018) · 3D Research · doi:10.1007/s13319-018-0203-x · Cited by 6
Minimization of Parallax Artifacts in Video Stitching for Moving Foregrounds
Kakli, M. U., Cho, Y., & Seo, J. (2018) · IEEE Access · doi:10.1109/access.2018.2871685 · Cited by 7
“The proposed scheme is tested with SIFT descriptor, SURF descriptor, and HLBP descriptor.”
InfluentialAccurate Sparse Feature Regression Forest Learning for Real-Time Camera Relocalization
Duong, N.-D., Kacete, A., Soladié, C., Richard, P.-Y., & Royan, J. (2018) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv.2018.00079 · Cited by 9
Real-time stitching method for infrared image
Cai, C., Fan, B., & Zhu, Q. (2018) · Optical Engineering · doi:10.1117/1.oe.57.11.113103 · Cited by 7
Lenslet Light Field Panorama Creation: A Sub-Aperture Image Stitching Approach
Oliveira, A., Brites, C., Ascenso, J., et al. (2018) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco.2018.8552948 · Cited by 7
“Feature Detection and Extraction: Local features are detected and extracted from all central sub-aperture images (one for each light field) using the SURF feature detector and extractor [11].”
Method3D Deep Object Recognition and Semantic Understanding for Visually-Guided Robotic Service
Lee, S., Naguib, A. M., & Islam, N. U. (2018) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8593985 · Cited by 7
“Many of the invariant photometric and geometric features developed in the past in 2D and 3D represent an attempt of achieving such dependability as described above [1-5].”
BackgroundHyperspectral band selection using structural information via hierarchical clustering
Delibaşoğlu, İ., & Çetin, M. (2019) · Journal of Applied Remote Sensing · doi:10.1117/1.jrs.13.014526 · Cited by 7
Robust Line Matching for Image Sequences Based on Point Correspondences and Line Mapping
Jia, X., Huang, X., Zhang, F., Gao, Y., & Yang, C. (2019) · IEEE Access · doi:10.1109/access.2019.2906368 · Cited by 7
“The first is the keypoint matching algorithm [11], [12], whose robustness and accuracy have been greatly improved in the past three decades.”
BackgroundWeakly Supervised Deep Metric Learning for Template Matching
Buniatyan, D., Popovych, S., Ih, D., Macrina, T., Zung, J., & Seung, H. S. (2019) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-17795-9_4 · Cited by 8
ARM-VO: an efficient monocular visual odometry for ground vehicles on ARM CPUs
Nejad, Z. Z., & Ahmadabadian, A. H. (2019) · Machine Vision and Applications · doi:10.1007/s00138-019-01037-5 · Cited by 9
“Small-sized binary descriptors such as BRIEF [23], ORB [24] and BRISK [25] which have been successfully used in ORB-SLAM2 [26] and OKVIS [27] allow a significant boost in matching compared to memory-hungry floating point descriptors such as SIFT [28] and SURF [29].”
MethodHese, S., Thiel, C., & Henkel, A. (2019) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xlii-2-w13-363-2019 · Cited by 7
Kim, J. (2019) · Symmetry · doi:10.3390/sym11070882 · Cited by 7
Deep Residual Texture Network for Terrain Recognition
Song, P., Ma, X., Li, X., & Li, Y. (2019) · IEEE Access · doi:10.1109/access.2019.2926994 · Cited by 8
Lee, S. J., & Hwang, S. S. (2019) · International Journal of Control Automation and Systems · doi:10.1007/s12555-018-0790-6 · Cited by 8
“[29] H. Bay, T. Tuytelaars, and L. Van Gool, “SURF: speeded up robust features,” Lect.”
BackgroundInfluential360° Video Stitching for Dual Fisheye Cameras
Lo, I.-C., Shih, K.-T., & Chen, H. H. (2019) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2019.8803804 · Cited by 7
“However, because of the use of traditional feature descriptors (such as SIFT [8] and SURF [9]) for image warping, their method is fraught with difficulties in dealing with fisheye distortion.”
MethodLearning Coarse and Fine Features for Precise Temporal Action Localization
Kim, J., & Heo, J. (2019) · IEEE Access · doi:10.1109/access.2019.2946898 · Cited by 10
“Improved dense trajectories (iDT) [21] introduced a warping algorithm to remove the global camera movement in the video based on RANSAC [22] with the SURF [23] features and optical flow.”
MethodIntegration of optimal spatial distributed tie-points in RANSAC-based image registration
Zhang, S., Li, S., Zhang, B., et al. (2020) · European Journal of Remote Sensing · doi:10.1080/22797254.2020.1724519 · Cited by 7
A GMS-Guided Approach for 2D Feature Correspondence Selection
Tang, Q., Yang, J., Jia, W., et al. (2020) · IEEE Access · doi:10.1109/access.2020.2975670 · Cited by 7
Fixed Pattern Noise Analysis for Feature Descriptors in CMOS APS Images
Zapata, J., Doménech‐Asensi, G., Ruı́z, R., et al. (2020) · Sensing and Imaging · doi:10.1007/s11220-020-0278-3 · Cited by 7
“On the one hand, there are floating-point descriptors based on the Histogram of Gradients (HOG), that is, SIFT [14], SURF [15], VGG [16] and Daisy [17].”
MethodScene image retrieval with siamese spatial attention pooling
Ma, J., & Gu, X. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.05.090 · Cited by 8
Mixed Reality: A Known Unknown
Sobota, B., Korečko, Š., Hudák, M., et al. (2020) · IntechOpen eBooks · doi:10.5772/intechopen.92827 · Cited by 7
Feature selection based on buzzard optimization algorithm for potato surface defects detection
Arshaghi, A., Ashourian, M., & Ghabeli, L. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09236-3 · Cited by 9
“The SURF algorithm was developed by Bay et al. [5]; this method was presented as a local feature identifier and descriptor which is inspired by the SIFT algorithm while enabling faster operation compared to the SIFT algorithm.”
MethodMallick, A. K., & Mukhopadhyay, S. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09312-8 · Cited by 7
“Taking this into consideration, we extract SURF [4] features whose dimension is empirically taken to be 64.”
MethodInfluentialChallenges in Face Recognition Using Machine Learning Algorithms: Case of Makeup and Occlusions
Selitskaya, N., Sielicki, S., & Christou, N. (2020) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-55187-2_9 · Cited by 7
A Novel Hardware–Software Co-Design and Implementation of the HOG Algorithm
Ghaffari, S., Soleimani, P., Li, K. F., & Capson, D. W. (2020) · Sensors · doi:10.3390/s20195655 · Cited by 7
Video Logo Retrieval Based on Local Features
Guan, B., Ye, H., Liu, H., et al. (2020) · Proceedings - International Conference on Image Processing · doi:10.1109/icip40778.2020.9191208 · Cited by 7
“Before 2012, Bag-of-Words (BoW) models with hand-crafted features were predominant, including descriptors such as SIFT, SURF and their variants [13, 14, 15, 5].”
MethodZhao, P., Hu, Q., Tang, Z., et al. (2020) · ISPRS International Journal of Geo-Information · doi:10.3390/ijgi9100596 · Cited by 7
Convolutional Neural Network based Deep Feature Extraction in Remote Sensing Images
Devi, P. D., & Thanuja, R. (2020) · International Conference on Smart Electronics and Communication (ICOSEC) · doi:10.1109/icosec49089.2020.9215274 · Cited by 7
“The propsoed method uses a deep feature extraction by using SURF algorithm [13].”
Method3D Reconstruction through Fusion of Cross-View Images
Qin, R., Song, S., Ling, X., & Elhashash, M. (2020) · IntechOpen eBooks · doi:10.5772/intechopen.93099 · Cited by 13
“Point features can be understood as corners or spots that are distinctive and easily identifiable across different images with various levels of perspective differences and typical features are SIFT (Scale-Invariant Feature Transform) [33], SURF (Speeded up robust features) [34], ORB (Oriented FAST and Rotated BRIEF)…”
BackgroundClassification of Fresh Cocoa Beans with Pulp Based on Computer Vision
Ona, A. J. O., Grijalva, F., Proano, K., et al. (2020) · IEEE ANDESCON · doi:10.1109/andescon50619.2020.9272188 · Cited by 7
“Specifically, we use the 64-dimensional SURF descriptor that is scale and rotation invariant [14].”
BackgroundMethodInfluentialEsparza, D. M. C., Terven, J. R., Romero-González, J.-A., et al. (2020) · Applied Sciences · doi:10.3390/app10248851 · Cited by 7
“…the same feature point in both images, for this, we compared corner detectors such as Harris [46] and Shi-Tomasi [47] along with feature point detectors such SIFT [48], SURF [49], BRISK [50], FAST [51], KAZE [52], a state-of-the-art Deep Learning method called SuperGlue [39] and DeepMatching [40].”
MethodDeep Learning Based Traffic Signs Boundary Estimation
Hrustic, E., Xu, Z., & Vivet, D. (2020) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv47402.2020.9304590 · Cited by 7
Performance Evaluation of Machine Learning Based Face Recognition Techniques
Sharma, S., & Kumar, V. (2021) · Wireless Personal Communications · doi:10.1007/s11277-021-08186-9 · Cited by 8
Three‐Dimensional Reconstruction of Tunnel Face Based on Multiple Images
Qiu, W., Jian, L., Cheng, Y., et al. (2021) · Advances in Civil Engineering · doi:10.1155/2021/8837309 · Cited by 7
Multimodal Urban Remote Sensing Image Registration Via Roadcross Triangular Feature
Yu, K., Zheng, X., Fang, B., et al. (2021) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2021.3073573 · Cited by 7
Drone Detection by Neural Network Using GLCM and SURF Features
Ahmed, T., Rahman, T., Roy, B. B., et al. (2021) · Journal of Information Systems and Telecommunication (JIST) · doi:10.52547/jist.9.33.15 · Cited by 7
“Finally, the descriptors are compared for matching among the images and the common points are taken as the matched attributes of the images [26, 28].”
BackgroundELMAS, B. (2021) · Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi · doi:10.17341/gazimmfd.689038 · Cited by 7
Interest Points as a Focus Measure in Multi-Spectral Imaging
Zukal, M., Mekyska, J., Cika, P., & Smekal, Z. (2013) · Brno University of Technology Digital Library (Brno University of Technology) · Cited by 7
A Case-Based Reasoning Model Powered by Deep Learning for Radiology Report Recommendation.
Amador-Domínguez, E., Serrano, E., Manrique, D., & Bajo, J. (2021) · International Journal of Interactive Multimedia and Artificial Intelligence · doi:10.9781/ijimai.2021.08.011 · Cited by 8
Digging Into Self-Supervised Learning of Feature Descriptors
Melekhov, I., Laskar, Z., Li, X., Wang, S., & Kannala, J. (2021) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv53792.2021.00122 · Cited by 10
Zhang, P., Luo, X., Ma, Y., Wang, C., Wang, W., & Xu, Q. (2022) · Remote Sensing · doi:10.3390/rs14030573 · Cited by 6
Low-level feature extraction (including edge detection)
Nixon, M. S., & Aguado, A. S. (2019) · Elsevier eBooks · doi:10.1016/b978-0-12-814976-8.00004-x · Cited by 8
Comparison of Modern Description Methods for the Recognition of 32 Plant Species
Kazerouni, M. F., Schlemper, J., & Kuhnert, K.-D. (2015) · Signal & Image Processing An International Journal · doi:10.5121/sipij.2015.6201 · Cited by 13
Automatic geolocation of targets tracked by aerial imaging platforms using satellite imagery
Shukla, P. K., Goel, S., Singh, P. P., & Lohani, B. (2014) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprsarchives-xl-1-381-2014 · Cited by 6
Leveraging High-Frequency Components for Deepfake Detection
Mejri, N., Papadopoulos, K., & Aouada, D. (2021) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp53017.2021.9733606 · Cited by 11
“The work in [16] uses speeded up robust features (SURF) [17] and support vector machines (SVM) to discriminate between real and swapped faces.”
MethodAppearance-based odometry and mapping with feature descriptors for underwater robots
Botelho, S. S. D. C., Drews, P., Figueiredo, M. D. S., et al. (2009) · Journal of the Brazilian Computer Society · doi:10.1590/s0104-65002009000300005 · Cited by 7
L2AMF-Net: An L2-Normed Attention and Multi-Scale Fusion Network for Lunar Image Patch Matching
Zhong, W., Jiang, J., & Ma, Y. (2022) · Remote Sensing · doi:10.3390/rs14205156 · Cited by 7
He, Z., Zhao, J., Zhao, X., Feng, W., Wang, Q., & Jiang, H. (2022) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-022-10333-w · Cited by 10
Exploring attention on faces: similarities between humans and Transformers
Cadoni, M. I., Nixon, S., Lagorio, A., & Fadda, M. (2022) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss56176.2022.9959705 · Cited by 7
“Later, more complex methods inspired by 978-1-6654-6382-9/22/$31.00 ©2022 IEEE human vision such as the Scale Invariant Feature Transform (SIFT) [30] and Speeded Up Robust Features (SURF) [3] found successful applications in object and face recognition [7, 22, 17, 28, 25].”
MethodInfluentialCentroid-based graph matching networks for planar object tracking
Li, K., Liu, H., & Wang, T. (2023) · Machine Vision and Applications · doi:10.1007/s00138-023-01382-6 · Cited by 7
COMPLEMENTARITY OF FEATURE POINT DETECTORS
Gales, G., Crouzil, A., & Chambon, S. (2010) · International Conference on Computer Vision Theory and Applications · doi:10.5220/0002831703340339 · Cited by 7
An improved tracking-learning-detection method
Wen, H., Wu, G., & Li, J. (2015) · Chinese Control Conference (CCC) · doi:10.1109/chicc.2015.7260234 · Cited by 7
CBRISK: Colored Binary Robust Invariant Scalable Keypoints
Jing, H., He, X. J., Han, Q., et al. (2013) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.e96.d.392 · Cited by 7
“Chu et al. [11] obtained the C-SURF descriptor according to the way producing C-SIFT, getting descriptor from the C color space consisting one intensity channel and 2 channels {Cλ,Cλλ}.”
MethodInfluential
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.