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 9,401–9,500 of 14,752
An efficient depth estimation using temporal 3D-Warping
Kumar, S. H., Suraj, K., & Ramakrishnan, K. R. (2014) · International Conference on D Imaging · doi:10.1109/ic3d.2014.7032586 · Cited by 2
“For SURF feature extraction and matching, we have used the built-in functions provided by MATLAB.”
MethodInfluentialMultiview Matching of Articulated Objects
Zini, L., Odone, F., & Cavallaro, A. (2014) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2014.2302547 · Cited by 2
“ture [17], and gradient location-orientation histogram [18].”
Discovery of Environmental Web Resources Based on the Combination of Multimedia Evidence
Tsikrika, T., Latas, A., Moumtzidou, A., Chatzilari, E., Vrochidis, S., & Kompatsiaris, I. Y. (2015) · doi:10.1145/2764873.2764876 · Cited by 2
“For the shallow representation, the popular and widely adopted SIFT [13] and SURF [1] descriptors are used.”
MethodKernel-Based On-Line Object Tracking Combining both Local Description and Global Representation
Miao, Q., Wang, G., & Lin, X. (2012) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.e96.d.159 · Cited by 2
Some improvements in the RGB-D SLAM system
Quang, H. P., & Quoc, N. L. (2015) · doi:10.1109/rivf.2015.7049884 · Cited by 2
“In this work, we choose SURF and ORB because of their fast computation time, which is suitable for real-time applications.”
MethodInfluentialScale and Rotation Robust Genetic Programming-Based Corner Detectors
Seo, K., & Kim, Y. (2010) · Journal of Institute of Control Robotics and Systems · doi:10.5302/j.icros.2010.16.4.339 · Cited by 2
A word-based approach for duplicate picture in picture sequence detection
Wang, L., Dong, Y., Bai, H., et al. (2011) · IEEE International Conference on Broadband Network and Multimedia Technology (IC-BNMT) · doi:10.1109/icbnmt.2011.6155942 · Cited by 2
Active stereoscopic camera to build an occupancy grid for autonomous navigation
Barcelos, A. D. O. P., Vidal, F. S., & Rosa, P. F. F. (2014) · IEEE International Symposium on Industrial Electronics (ISIE) · doi:10.1109/isie.2014.6864778 · Cited by 2
“Other computational vision algorithms can also be used, such as SURF (Speeded up Robust Features) [12], using a vector of 128 descriptors, which are faster than the SIFT algorithm, another algorithm presented in [13], ORB (Oriented Fast and Rotated Brief) is faster than SIFT”
MethodInfluentialA quantitative evaluation of the conceptual consistency of visual words and visual vocabularies
Stommel, M., Herzog, O., & Xu, W. (2014) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2014.11.015 · Cited by 2
Ontology-based 3D pose estimation for autonomous object manipulation
Kouskouridas, R., Retzepi, T., Charalampoglou, E., et al. (2012) · IEEE International Conference on Imaging Systems and Techniques (IST) · doi:10.1109/ist.2012.6295586 · Cited by 2
“The proposed technique was tested on the two most common two-part algorithms, namely the SIFT and the SURF [26], and was found to outperform with the first one.”
MethodReigSAC: fast discrimination of spurious keypoint correspondences on planar surfaces
Proença, H. (2014) · Machine Vision and Applications · doi:10.1007/s00138-014-0593-6 · Cited by 2
Springer, D., Frank, M., Simmet, F., et al. (2013) · Picture Coding Symposium · doi:10.1109/pcs.2013.6737718 · Cited by 2
“Much work regarding feature transform speedup has been dedicated to the use of fast descriptor generation and matching [7] [8] as well as to layer-based processing [6], which can be seen as a local search space restriction for lower layers.”
BackgroundLocation identification for visitor behavior log in museum
Aizawa, K., Yamasaki, T., Kawaji, H., et al. (2010) · doi:10.1145/1900179.1900256 · Cited by 2
Improved SURF algorithm based on SVM classification
Chang, J., Wei, W., & Liang, J. (2011) · Cited by 2
Huang, W., Luo, A.-C., Chen, W., et al. (2012) · 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2012.6238902 · Cited by 2
“These tools utilize feature extraction and matching algorithms similar to SIFT [7] or SURF [8] to estimate disparity distribution of stereoscopic images.”
MethodFusing mixed reality and networked home techniques to improve user control of consumer electronics
Belimpasakis, P., & Walsh, R. (2011) · International Conference on Consumer Electronics · doi:10.1109/icce.2011.5722487 · Cited by 2
“Techniques such as the SURF (Speeded Up Robust Features) [5], which is a robust image detector & descriptor, are utilized for identifying the object.”
MethodFeature-Based Detection of Inverted-Stereo for Stereoscopic 3D Viewing Comfort
Kim, W.-J., Leigh, S.-W., Hur, N., et al. (2012) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2012.2190193 · Cited by 2
Williams, H. A., Browne, W. N., & Milford, M. J. (2012) · QUT ePrints (Queensland University of Technology) · Cited by 2
Viewpoint invariant model for face detection
Taffar, M., Miguet, S., & Benmohammed, M. (2013) · International Journal of Computational Vision and Robotics · doi:10.1504/ijcvr.2013.056039 · Cited by 2
“The methods for local invariant features extraction (Lowe, 2004; Herbert et al., 2006) have been widely experimented in several publications (Hartley and Zisserman, 2000; Mikolajczyk and Schmid, 2004; Fergus et al.”
MethodRelative Relevance Feedback in Image Retrieval
Sugiyama, Y., Kato, M. P., Ohshima, H., & Tanaka, K. (2012) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2012.161 · Cited by 2
Visual odometry with high resolution time-of-flight cameras
Dalbah, Y., Dingeldey, N., & Wahl, F. M. (2013) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2050903 · Cited by 2
“For this task the descriptors SIFT and SURF respectively, have proven high discriminative power and high robustness [12].”
MethodAutomatic annotation extension and classification of documents using a probabilistic graphical model
Bouzaieni, A., Barrat, S., & Tabbone, S. (2015) · Proceedings of the International Conference on Document Analysis and Recognition · doi:10.1109/icdar.2015.7333775 · Cited by 2
Towards automatic stereo pair extraction for 3D visualisation of historical aerial photographs
Hast, A., & Marchetti, A. (2014) · International Conference on D Imaging · doi:10.1109/ic3d.2014.7032575 · Cited by 2
RGBD SLAM for Indoor Environment
Lin, R., Wang, Y., & Yang, S. (2013) · Advances in intelligent systems and computing · doi:10.1007/978-3-642-37835-5_15 · Cited by 2
“SIFT [8] and SURF [9] features are scale and viewpoint invariant but computationally demanding.”
BackgroundHigh-Speed and Local-Changes Invariant Image Matching
Zhang, C., & Akashi, T. (2015) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2015edp7093 · Cited by 2
Learning to Reconstruct 3D Structure from Object Motion
Liu, W., Dou, H., & Wu, X. H. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-26532-2_15 · Cited by 2
Global Object Representation of Scene Surveillance Video Based on Model and Feature Parameters
Ma, M., Hu, R., Chen, S., et al. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24075-6_22 · Cited by 2
Direct view manipulation for drone photography
Chen, Y., Lee, W., Chan, L., et al. (2015) · SIGGRAPH Asia Posters · doi:10.1145/2820926.2820945 · Cited by 2
“Speci.cally, we employ SURF feature matching [Bay et al. 2006] to evaluate the registration error between the current and target view by computing the displacement of the matched keypoints.”
IDENTIFYING PAINTINGS IN MUSEUM GALLERIES USING CAMERA MOBILE PHONES
Ruf, B., & Detyniecki, M. (2009) · doi:10.1142/9789814277563_0013 · Cited by 2
Drishti: An Ultra-Low Cost Visual-Aural Assistive Technology for the Visually Impaired
Kapur, A., & Kapur, S. (2014) · International Convention on Rehabilitation Engineering & Assistive Technology · Cited by 2
Automatic Motion Classification for Advanced Driver Assistance Systems
Desai, A., Lee, D.-J., & Mody, S. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-27863-6_77 · Cited by 2
STEREO RECONSTRUCTION OF ATMOSPHERIC CLOUD SURFACES FROM FISH-EYE CAMERA IMAGES
Kátai-Urbán, G., Otte, V., Kees, N., Megyesi, Z., & Bixel, P. (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-b3-49-2016 · Cited by 2
Large-scale image annotation with image–text hybrid learning models
Chien, B.-C., & Ku, C.-W. (2016) · Soft Computing · doi:10.1007/s00500-016-2221-z · Cited by 2
“The region-based segmentation considers visual features of images, such as colors, shapes and texture (Bay et al. 2006; Chatzichristofis and Boutalis 2008b; Comaniciu and Meer 1997; Feng et al. 2001; Fukui et al. 2011; Ma and Majunath 1997; Shi and Malik 2000).”
BackgroundA research-driven approach to undergraduate robotics education
Singh, S. P. N., Fitch, R., & Williams, S. B. (2010) · Cited by 2
Handheld pose tracking using vision-inertial sensors with occlusion handling
Li, J., Slembrouck, M., Deboeverie, F., et al. (2016) · Journal of Electronic Imaging · doi:10.1117/1.jei.25.4.041012 · Cited by 2
“EQ-TARGET;temp:intralink-;e020;326;591 P 1⁄4 A1⁄2X; Y; Z T 1⁄4 b; (20)”
Feature detection in biological tissues using multi-band and narrow-band imaging
Tamura, Y., Mashita, T., Kuroda, Y., et al. (2016) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-016-1458-4 · Cited by 2
AN INTEGRATIVE OBJECT-BASED IMAGE ANALYSIS WORKFLOW FOR UAV IMAGES
Yu, H., Yan, T., Yang, W. T., 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-b1-1085-2016 · Cited by 2
Visual Positioning in a Smartphone
Ruotsalainen, L., & Kuusniemi, H. (2012) · IGI Global eBooks · doi:10.4018/978-1-4666-1827-5.ch006 · Cited by 2
Image categorization with resource constraints: introduction, challenges and advances
Luo, J.-H., Zhou, W., & Wu, J. (2017) · Frontiers of Computer Science · doi:10.1007/s11704-016-5514-6 · Cited by 3
Structural iMoSIFT for human action recognition
Chen, H., Chen, J., & Hu, R. (2016) · Wuhan University Journal of Natural Sciences · doi:10.1007/s11859-016-1169-2 · Cited by 2
Application of machine learning to classify surface marker of screw
Chang, C., & Shie, H.-C. (2016) · IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan) · doi:10.1109/icce-tw.2016.7520948 · Cited by 2
Video similarity search by using compact representations
da Silva, H. B., Almeida, R., da Fonseca, G. B., Caetano, C., Vieira, D., Patrocínio, Z. K. G., Araújo, A. D. A., & Guimarães, S. J. F. (2016) · ACM Symposium on Applied Computing · doi:10.1145/2851613.2851876 · Cited by 2
Searching for TELUGU Script in Noisy Images Using SURF descriptors
Lakshmi, K. M., & Babu, T. R. (2016) · IEEE International Conference on Advanced Computing (IACC) · doi:10.1109/iacc.2016.95 · Cited by 2
“SURF feature transform extracts the unique feature descriptors to represent an image and also outperforms the state of art methods in this area [7].”
BackgroundMethodInfluentialImage retrieval with the use of Bag of Words and structural analysis
Malashin, R. O. (2016) · Journal of Physics Conference Series · doi:10.1088/1742-6596/735/1/012004 · Cited by 2
A statistical approach to rank the matched image points
Famouri, M., & Azimifar, Z. (2016) · Iranian Conference on Electrical Engineering · doi:10.1109/iraniancee.2016.7585706 · Cited by 2
“Most of the approaches tried to introduce a new image descriptor like SIFT [2], SURF [3], BRISK [4], BREAF [5] and ASIFT [6].”
MethodCombining depth and gray images for fast 3D object recognition
Wang, P., Zhu, F., & Hao, Y. (2016) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2247370 · Cited by 2
“And then create discriminative descriptors derived from the surroundings of the feature points[8][9].”
BackgroundRe‐Compositable Panoramic Selfie with Robust Multi‐Frame Segmentation and Stitching
Li, K., Wang, J., Liu, Y., et al. (2016) · Computer Graphics Forum · doi:10.1111/cgf.13020 · Cited by 2
“Traditional feature detectors, such as SIFT [Low99] and SURF [BTVG06], are designed to be robust against image transformations for matching and tracking.”
MethodAnalysis of visual correlation-extreme methods of UAV navigation
Mukhina, M. P., & Demchenko, D. (2013) · doi:10.1109/apuavd.2013.6705329 · Cited by 2
“Description of SURF method SURF-detector was first introduced in 2006 [3].”
MethodImproved simultaneous localization and mapping by stereo camera and SURF
Wang, T.-C., & Chen, C. (2013) · doi:10.1109/cacs.2013.6734133 · Cited by 2
“In section III and IV, we introduce how we use Speeded-Up Robust Features (SURF)[15] to detect and match feature points.”
MethodComputer Vision Performance and Image Quality Metrics : A Reciprocal Relation
Haccius, C., & Herfet, T. (2017) · doi:10.5121/csit.2017.70104 · Cited by 2
Tasaki, T., Tokura, S., Sonoura, T., Ozaki, F., & Matsuhira, N. (2011) · Journal of Robotics and Mechatronics · doi:10.20965/jrm.2011.p1012 · Cited by 2
Environment characterization using Laplace eigenvalues
Mihankhah, E., & Wang, D. (2016) · International Conference on Control, Automation, Robotics and Vision (ICARCV) · doi:10.1109/icarcv.2016.7838769 · Cited by 2
“Another feature based descriptor is introduced in [8] which is known as SURF.”
Learning deep similarity in fundus photography
Chudzik, P., Al-Diri, B., Calivá, F., et al. (2017) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2254286 · Cited by 2
“The need to compare images has driven the development of many highly successful hand-engineered feature descriptors like SIFT(1) and SURF.(2) Unfortunately, methods based on human-crafted features are limited in their modelling and generalization abilities by human ingenuity.”
MethodInfluentialUma Arquitetura de Software para Bibliotecas de Realidade Aumentada Baseada em Marcadores Naturais
de Oliveira, D. C. B., Pereira, L. T., & da Silva, R. L. D. S. (2017) · Revista de Informática Teórica e Aplicada · doi:10.22456/2175-2745.69479
“Outros extratores famosos como o SIFT (Scale-Invariant Feature Transform) [16] e o SURF (Speeded Up Robust Features) [17] obtêm uma qualidade superior, mas deixam muito a desejar no desempenho, além de serem patenteados.”
BackgroundA Logo-Based Approach for Recognising Multiple Products on a Shelf
Mittal, T., Laasya, B. M., & Babu, J. D. (2017) · Lecture notes in networks and systems · doi:10.1007/978-3-319-56991-8_2 · Cited by 2
RESEARCH OF EFFECTIVENESS OF STRUCTURAL IMAGE CLASSIFICATION METHODS USING CLUSTER DATA MODEL
Gorokhovatsky, V. A., Putyatin, Y. P., & Stolyarov, V. S. (2017) · Radio Electronics Computer Science Control · doi:10.15588/1607-3274-2017-3-9 · Cited by 2
“Множество СП в виде совокупности числовых векторов формируется применением специальных детекторов, наиболее известными среди которых в настоящее время считают SURF и более быстрый при реализации, но менее устойчивый к масштабированию, ORB [3, 5, 7].”
Score Fusion of SIFT & SURF Descriptors for Face Recognition Using Wavelet Transforms
Ameen, M. M., & Eleyan, A. (2017) · International Journal of Image Graphics and Signal Processing · doi:10.5815/ijigsp.2017.10.03 · Cited by 2
Aoki, R., Aoki, S., Ohtagaki, Y., et al. (2017) · IEEE International Conference on Consumer Electronics - Berlin (ICCE-Berlin) · doi:10.1109/icce-berlin.2017.8210589 · Cited by 2
“However, for 2D illustrations that contains many regions with only a solid color, SURF is expected to improve accuracy because it adopts box filters to compute gradients.”
MethodInfluentialUltrasonic Testing of Manufactured Voids in Electron Beam Malted Ti-6Al-4V
Durkee, A. D., Lingenfelter, A. J., & Liu, D. (2018) · AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference (SDM) · doi:10.2514/6.2018-0233 · Cited by 2
Design of UAV distributed aided navigation simulation system based on scene/terrain matching
Cheng, S., Wang, H., Ji, H., et al. (2017) · Asian Control Conference (ASCC) · doi:10.1109/ascc.2017.8287196 · Cited by 2
“The work of SURF [9] scene matching is as follows.”
MethodInfluentialHigh-Speed Recognition Algorithm Based on BRISK and Saliency Detection for Aerial Images
Teng-jiao, X., Zhao, D., Shi, J. T., et al. (2013) · Research Journal of Applied Sciences Engineering and Technology · doi:10.19026/rjaset.5.4221 · Cited by 2
Loop Closure Detection Using Local Invariant Features and Randomized KD-Trees
Garcia‐Fidalgo, E., & Ortiz, A. (2018) · Springer tracts in advanced robotics · doi:10.1007/978-3-319-75993-7_5 · Cited by 2
Real-time Detection and Tracking Method of Landmark Based on UAV Visual Navigation
Li, J., Ma, X., Chen, H., Duan, X., & Zhang, Y. (2018) · Xibei Gongye Daxue Xuebao/Journal of Northwestern Polytechnical University · doi:10.1051/jnwpu/20183620294 · Cited by 2
Visualization of the Past-to-Recent Changes in Cultural Heritage Based on 3D Digitization
Mori, N., Higo, T., Suemori, K., et al. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01762-0_1 · Cited by 2
FEATURE VECTOR GENERATION FOR THE FACIAL EXPRESSION RECOGNITION USING NEO-FUZZY SYSTEM
Bodyanskiy, Y. V., Kulishova, N. Y., & Tkachenko, V. (2018) · Radio Electronics Computer Science Control · doi:10.15588/1607-3274-2018-3-10 · Cited by 2
Online Inertial-Aided Monocular Camera Self-Calibration
Pio, A. B., & Borges, G. A. (2018) · Latin American Robotics Symposium, Brazilian Symposium on Robotics and Workshop on Robotics in Education (LARS/SBR/WRE) · doi:10.1109/lars/sbr/wre.2018.00027 · Cited by 2
Practical Algorithm Design for the FFT-based Robust Image Registration Method
Ri, Y., & Fujimoto, H. (2018) · IEEJ Transactions on Industry Applications · doi:10.1541/ieejias.139.22 · Cited by 2
A Real-time Fusion System for Multi-focus Microscopic Images Based on Mobile Platform
Ding, W., Lai, T., Zhu, C., et al. (2019) · IEEE Joint International Information Technology and Artificial Intelligence Conference (ITAIC) · doi:10.1109/itaic.2019.8785591 · Cited by 2
“Benefited with the relatively good robustness, noise immunity and the fast operation speed, we select a SURF feature-based algorithm for correction.”
MethodInfluentialGeneración eficiente de mosaicos para inspección de infraestructuras lineales mediante robots aéreos
Tapia, R., Dios, J. R. M., & Baturone, A. O. (2019) · Jornadas de Automática · doi:10.17979/spudc.9788497497169.802 · Cited by 2
Improved-ORB-based SLAM Using RGB-D Data
Zhou, Y., Li, B., Li, Y., & Wang, D. (2019) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc.2019.8832804 · Cited by 2
“To ensure the real-time performance of the VSLAM system, SIFT (-300 ms) [16], SURF (-300 ms) [17], and A-KAZE (100 m s)[18], which are popular in feature selection, cannot be selected.”
MethodContact Lens Detection for Security
Cayabyab, G. T. (2019) · International Journal for Research in Applied Science and Engineering Technology · doi:10.22214/ijraset.2019.5282 · Cited by 2
“…Group Matching algorithm used descriptors to detect a set of key points in the image and provide a description on vectors which is Scale Invariant Feature Transform (SIFT) [26], [27] and Speeded Up Robust Features (SURF) [28] algorithm which represent a starting point for the FGM elaboration.”
MethodWen, S., Wang, S., Zhang, Z., Zhang, X., & Zhang, D. (2019) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2019.00105 · Cited by 2
“One is SURF feature extraction (Bay et al., 2006).”
Multimodal registration of visible, SWIR and LWIR images in a distributed smart camera system
Cárdenas, J., Soto, J. E., & Figueroa, M. E. (2019) · Microprocessors and Microsystems · doi:10.1016/j.micpro.2019.102987 · Cited by 2
3D Pose Estimation and Tracking of an Electricity Pylon
Ali, E. Y., & Nicolls, F. (2020) · 2020 International SAUPEC/RobMech/PRASA Conference · doi:10.1109/saupec/robmech/prasa48453.2020.9041094 · Cited by 2
“Feature-based techniques such as SIFT [6], SURF [7] and ORB [8] work well with natural images that are well textured.”
MethodPOP: A Generic Framework for Real-Time Pose Estimation of Planar Objects
Chae, S., Hong, J. H., Choi, H., et al. (2020) · IEEE Access · doi:10.1109/access.2020.3020309 · Cited by 2
“Such descriptors include binary robust independent elementary features (BRIEF) [8], binary robust invariant scalable key-points (BRISK) [29], speeded-up robust features (SURF) [3].”
BackgroundMethodA Fusion of Visible and Infrared Images for Victim Detection
Gupta, M. (2020) · Studies in computational intelligence · doi:10.1007/978-981-15-6844-2_12 · Cited by 2
A Novel Key Generation Approach Based on Facial Image Features for Stream Cipher System
Sannidhan, M. S., Sudeepa, K. B., Martis, J. E., et al. (2020) · International Conference on Smart Systems and Inventive Technology (ICSSIT) · doi:10.1109/icssit48917.2020.9214095 · Cited by 2
“Several research studies have claimed that the extractor is faster as well as robust for any transformations on the image and is applied in the area of object detection, object tracking, facial recognition etc. [14,15].”
BackgroundPerformance Prediction for Multi-Application Concurrency on GPUs
Moolchandani, D., Gupta, S., Kumar, A., et al. (2020) · IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) · doi:10.1109/ispass48437.2020.00050 · Cited by 2
“Sift Extracts those features from an image that are invariant to image orientation, illumination and scaling Surf Feature extraction [33] with scale invariance.”
BackgroundA fast and robust real-time surveillance video stitching method
Yang, T., Jin, F., & Luo, J. (2020) · Journal of Physics Conference Series · doi:10.1088/1742-6596/1651/1/012170 · Cited by 2
“Specifically, we use the classical combination of SURF [13] feature detecting, 2-NN matching, RANSAC [14] purification and cylindrical projection in image registration.”
MethodKong, W., Zang, P., Niu, S., & Li, D. (2021) · IET Image Processing · doi:10.1049/ipr2.12054 · Cited by 2
Dense Point Cloud Mapping Based on RGB-D Camera in Dynamic Indoor Environment
Zhang, F., Li, Q., Wang, T., & Liu, Y. (2020) · Chinese Automation Congress · doi:10.1109/cac51589.2020.9327342 · Cited by 3
“However, until now, most of the discussion has been about localization, including feature point localization [1] [2] [3], direct localization [4] [5], and bundle adjustment (BA) optimization [6].”
BackgroundA fast stitching method for container images using texture and weighted speed
Meng, Q., Zhang, M., & Zhang, W. (2019) · doi:10.4108/eai.29-6-2019.2283071 · Cited by 2
“After that, SURF[6] is used to detect keypoints and FLANN[7] is used to match those keypoints.”
MethodInfluentialAdditive margin cosine loss for image registration
Ma, Y., Sun, S., Wu, F., Yang, Y., Yang, X., Xu, B., & Luo, Z. (2021) · The Visual Computer · doi:10.1007/s00371-021-02105-6 · Cited by 2
“Representative among them is Harris algorithm, which compares the change rate of strength in two orthogonal directions [1], SIFT [2] algorithm, principal component analysis sift (PCA-SIFT) [3] algorithm, and speed up robust feature (SURF) [4] algorithm, which is compared by extracting key points.”
MethodApplication of Learning-based Multi-view Stereo in Digital Design
Zhang, K., & Liu, M. (2021) · International Conference on Electronics, Circuits and Information Engineering (ECIE) · doi:10.1109/ecie52353.2021.00075 · Cited by 2
Generation of Stereo Images from the Heterogeneous Cameras
Kim, S. (2021) · Instrumentation Mesure Métrologie · doi:10.18280/i2m.200202 · Cited by 2
“To find feature points, SIFT (Scale Invariant Feature Transformation) [12], SURF (Speeded Up Robust Features) [13] and ORB (Oriented and Rotated BRIEF) [14] algorithms can be used.”
MethodResearch on Household Waste Detection System Based on Deep Learning
Li, M., & Chengrong, L. (2021) · E3S Web of Conferences · doi:10.1051/e3sconf/202126701034 · Cited by 2
Automatic visual inspection system for solder paste deposition in PCBs
Paulo, G. B. F., Perdigoto, L., & de Faria, S. M. M. (2021) · doi:10.1109/conftele50222.2021.9435522 · Cited by 2
Total Least Squares for Optimal Pose Estimation
Maleki, S., Crassidis, J. L., Cheng, Y., & Schmid, M. (2022) · AIAA SciTech Forum · doi:10.2514/6.2022-1222 · Cited by 4
“A class of stable algorithms exist for this purpose, such as SIFT [19] and SURF [20].”
MethodResidual Vector Capsule: Improving Capsule by Pose Attention
Xie, N., & Wan, X. (2021) · IEEE Access · doi:10.1109/access.2021.3113176 · Cited by 2
Hafez, A. H. A., Tello, A., & Alqaraleh, S. (2021) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-021-06148-8 · Cited by 3
“FabMap work presented in [2] used the SURF features [6], while SeqS-LAM [1] has used the downsized and contrast normalized images as features for matching purpose.”
MethodPose Estimation Algorithm for Helicopter Landing Based on YOLO and PNP
Zhou, R., She, J., Qi, N., Yu, L., & Liu, Y. (2021) · Lecture notes in electrical engineering · doi:10.1007/978-981-15-8155-7_252 · Cited by 3
Content Based Video Retrieval—Methods, Techniques and Applications
Mounika, B., Palanisamy, P., & Khare, A. (2021) · Studies in big data · doi:10.1007/978-3-030-75657-4_4 · Cited by 2
A Robust Invariant Local Feature Matching Method for Changing Scenes
Wang, D., Zhang, H., & Shao, Y. (2021) · Wireless Communications and Mobile Computing · doi:10.1155/2021/8927822 · Cited by 2
A Feature Extraction Approach Based on LBP Operator and Complex Networks for Face Recognition
Piotto, J. G. D. S., & Lopes, F. M. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-93420-0_41 · Cited by 2
Automated Computer Vision-Based Reading of Residential Meters
Koščević, K., & Subašić, M. (2016) · Proceedings of the Croatian Computer Vision Workshop · doi:10.20532/ccvw.2016.0015 · Cited by 2
A Vision based Vehicle Detection System
Chandel, H., & Vatta, S. (2015) · Communications on Applied Electronics · doi:10.5120/cae2015651767 · Cited by 2
Pitchandi, P. (2022) · Advances in Science and Technology – Research Journal · doi:10.12913/22998624/150320 · Cited by 2
Image retargeting quality assessment: A survey
Guo, Y., Wang, D., Yan, G., et al. (2022) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-220456 · Cited by 2
Stabilization of spherical videos based on feature uncertainty
Luchetti, A., Zanetti, M., Kalkofen, D., & Cecco, M. D. (2022) · The Visual Computer · doi:10.1007/s00371-022-02578-z · Cited by 5
TG: Accurate and Efficient RGB-D Feature With Texture and Geometric Information
Zhao, X., Liu, Y., Wang, Z., et al. (2022) · IEEE/ASME Transactions on Mechatronics · doi:10.1109/tmech.2022.3175812 · Cited by 2
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.