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 3,401–3,500 of 14,752
Li, Z., Zhang, Z., Chen, D., et al. (2022) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2022.3188515 · Cited by 19
SMDAF: A novel keypoint based method for copy‐move forgery detection
Yue, G., Duan, Q., Liu, R., Peng, W., Liao, Y., & Liu, J. (2022) · IET Image Processing · doi:10.1049/ipr2.12578 · Cited by 21
Evolving graph-based video crowd anomaly detection
Yang, M., Feng, Y., Rao, A. S., Rajasegarar, S., Tian, S., & Zhou, Z. (2023) · The Visual Computer · doi:10.1007/s00371-023-02783-4 · Cited by 20
“As SURF is suitable for illumination changes in images [26] and can be more efficient in the feature selection stage through a fast approximation of the Hessian when compared with SIFT, we choose SURF as the feature descriptor in this stage.”
Voice User Interface: Literature review, challenges and future directions
Francis, R., Randriatsarafara, H. N., Hajalalaina, A. R., & Manda-Vy, R. N. (2021) · SYSTEM THEORY CONTROL AND COMPUTING JOURNAL · doi:10.52846/stccj.2021.1.2.26 · Cited by 32
ParaFormer: Parallel Attention Transformer for Efficient Feature Matching
Lu, X., Yan, Y., Kang, B., & Ду, С. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i2.25275 · Cited by 30
Analysis of recent techniques in marine object detection: a review
Bhalla, S., Kumar, A., & Kushwaha, R. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-19782-9 · Cited by 20
Towards persistent visual navigation using SMART
Pepperell, E., Corke, P., & Milford, M. J. (2013) · QUT ePrints (Queensland University of Technology) · Cited by 22
Vehicle Classification Based on Soft Computing Algorithms
Dalka, P., & Czyżewski, A. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-13529-3_9 · Cited by 23
Comparison of feature detection techniques for AUV navigation along a trained route
King, P., Anstey, B., & Vardy, A. (2013) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.23919/oceans.2013.6741023 · Cited by 19
“The feature types tested include SIFT[12], SURF[3], MSER[13], STAR[1], ORB[15], and BRIEF[4].”
MethodOMG: Introducing optical myography as a new human machine interface for hand amputees
Nissler, C., Mouriki, N., Castellini, C., Belagiannis, V., & Navab, N. (2015) · International Conference on Rehabilitation Robotics · doi:10.1109/icorr.2015.7281324 · Cited by 24
Mid-level image representations for real-time heart view plane classification of echocardiograms
Penatti, O. A. B., Werneck, R. D. O., de Almeida, W. R., Stein, B. V., Pazinato, D. V., Júnior, P. R. M., Torres, R. D. S., & Rocha, A. D. R. (2015) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2015.08.004 · Cited by 20
“Although SIFT was used, we believe that the impact of using similar descriptors, like Speeded Up Robust Features (SURF) [35] or others alike, is minimum.”
MethodImage Flower Recognition based on a New Method for Color Feature Extraction
Mabrouk, A. B., Najjar, A., & Zagrouba, E. (2014) · International Conference on Computer Vision Theory and Applications · doi:10.5220/0004636302010206 · Cited by 21
Tiwari, S., & Singh, S. K. (2012) · IET Biometrics · doi:10.1049/iet-bmt.2012.0040 · Cited by 19
HDR IMAGING FOR FEATURE DETECTION ON DETAILED ARCHITECTURAL SCENES
Kontogianni, G., Stathopoulou, E. K., Georgopoulos, A., & Doulamis, A. D. (2015) · 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-5-w4-325-2015 · Cited by 24
“Bay et al. (Bay et al 2006) presented an algorithm named SURF (Speed Up Robust Features) that outperformed the existing state of the art detectors and descriptors in terms of repeatability, distinctiveness and robustness as well as speed.”
MethodInfluentialChoi, J. H., Park, Y. W., Song, J.-B., et al. (2011) · International Journal of Control Automation and Systems · doi:10.1007/s12555-011-0413-y · Cited by 19
“A vocabulary tree [6] with SURF (Speeded Up Robust Feature) [7] is used for fast and efficient matching and the searching process in visual localization.”
Method3D Face Recognition using Kinect
Ajmera, R., Nigam, A., & Gupta, P. (2014) · Indian Conference on Computer Vision, Graphics and Image Processing · doi:10.1145/2683483.2683559 · Cited by 19
Dibs, H., Mansor, S. B., Ahmad, N., & Pradhan, B. (2015) · International Journal of Remote Sensing · doi:10.1080/01431161.2015.1034891 · Cited by 20
“These algorithms include Speeded-Up Robust Features (SURF) by Bay, Tuytelaars, and Gool (2006) and the Gradient Location and Orientation Histogram by Mikolajczyk and Schmid (2005).”
On-Road Collision Warning Based on Multiple FOE Segmentation Using a Dashboard Camera
Yang, M., & Zheng, J. (2015) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2014.2378373 · Cited by 20
“Theoretically, SIFT and SURF are invariant to scale and should track better than GFTT.”
MethodInfluentialAutomated computation of the fundamental matrix for vision based construction site applications
Jog, G. M., Fathi, H., & Brilakis, I. (2011) · Advanced Engineering Informatics · doi:10.1016/j.aei.2011.03.005 · Cited by 19
“As an alternative, various automatic point detection methods, such as Harris corner detector [9], Kanade-Lucas-Tomasi (KLT) feature tracker [10], Scale Invariant Robust Features (SIFT) detector [11], and Speeded Up Robust Features (SURF) detector [12], have been used to find distinctive features of the scene.”
MethodReal time static hand gesture recognition system prototype for Indonesian sign language
Hartanto, R., Susanto, A., & Santosa, P. I. (2014) · doi:10.1109/iciteed.2014.7007911 · Cited by 19
Haner, S., & Gu, I. Y. (2010) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2010.1112 · Cited by 19
A probabilistic framework for fusing frame-based searches within a video copy detection system
Gengembre, N., & Berrani, S.-A. (2008) · doi:10.1145/1386352.1386384 · Cited by 19
“Interest points are detected in the keyframes by a fast Hessian detector as proposed by [3].”
Online environment mapping using metric-topological maps
Lim, J., Frahm, J., & Pollefeys, M. (2012) · The International Journal of Robotics Research · doi:10.1177/0278364912461455 · Cited by 21
Efficient Path-Based Stereo Matching With Subpixel Accuracy
Donate, A., Liu, X., & Collins, E. G. (2010) · IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics) · doi:10.1109/tsmcb.2010.2049839 · Cited by 20
FIND: A Neat Flip Invariant Descriptor
Guo, X., & Cao, X. (2010) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2010.131 · Cited by 19
“Most of the content-based methods are based on local image descriptors, such as SIFT [7] and SURF [1], to extract image features as content information.”
MethodShape-from-recognition: Recognition enables meta-data transfer
Thomas, A., Ferrari, V., Leibe, B., Tuytelaars, T., & Gool, L. V. (2009) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2009.03.010 · Cited by 18
Self-citationA novel local mesh color texture pattern for image retrieval system
Yesubai Rubavathi, C., & Ramraj, R. (2015) · AEU - International Journal of Electronics and Communications · doi:10.1016/j.aeue.2015.11.009 · Cited by 19
Component-Wise Modeling of Articulated Objects
Ntouskos, V., Sanzari, M., Cafaro, B., Nardi, F., Natola, F., Pirri, F., & Ruiz, M. (2015) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.268 · Cited by 19
“by keypoints, SURF [2] features or similar); 4 Project Fic on Bic to obtain the 3D feature points Xic; 5 Find feature matches Fic ↔ Frc; 6 if #matches > 3 then”
BackgroundA mosaicing approach for vessel visual inspection using a micro-aerial vehicle
Garcia‐Fidalgo, E., Ortiz, A., Bonnin‐Pascual, F., & Joan P. Company (2015) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2015.7353361 · Cited by 20
A maximally stable extremal regions system-on-chip for real-time visual surveillance
Salahat, E. N., Saleh, H. H., Sluzek, A. S., Al‐Qutayri, M., Mohammad, B. S., & Ismail, M. (2015) · Annual Conference of the IEEE Industrial Electronics Society (IECON) · doi:10.1109/iecon.2015.7392528 · Cited by 19
Affine versus projective transformation for SIFT and RANSAC image matching methods
Redzuwan, R. M., Radzi, N. A. M., Din, N. M., et al. (2015) · IEEE International Conference on Signal and Image Processing Applications (ICSIPA) · doi:10.1109/icsipa.2015.7412233 · Cited by 19
“Keywords—SIFT; SURF; RANSAC; affine transformation; projective transformation I. INTRODUCTION The first process that needs to be done to produce a panoramic image is image registration.”
MethodNoise robust and rotation invariant entropy features for texture classification
Shakoor, M. H., & Tajeripour, F. (2016) · Multimedia Tools and Applications · doi:10.1007/s11042-016-3455-6 · Cited by 19
Fast depth edge detection and edge based RGB-D SLAM
Bose, L., & Richards, A. G. (2016) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2016.7487265 · Cited by 19
“Many monocular camera based SLAM systems ([3], [4], [5]) utilize visual point features such as the well known SIFT [6] and SURF [7] feature descriptors.”
MethodWang, G., Wang, Z., Chen, Y., Zhou, Q., & Zhao, W. (2016) · Information Sciences · doi:10.1016/j.ins.2016.08.041 · Cited by 19
FPGA-based parallel hardware architecture for SIFT algorithm
Peng, J., Liu, Y. H., Lyu, C., Li, Y., Zhou, W., & Fan, K. (2016) · IEEE International Conference on Real-time Computing and Robotics (RCAR) · doi:10.1109/rcar.2016.7784039 · Cited by 19
INFLUENCE OF BLUR ON FEATURE MATCHING AND A GEOMETRIC APPROACH FOR PHOTOGRAMMETRIC DEBLURRING
Sieberth, T., Wackrow, R., & Chandler, J. H. (2015) · Cited by 19
Efficient motion estimation methods for fast recognition of activities of daily living
Poularakis, S., Avgerinakis, K., Briassouli, A., & Kompatsiaris, I. Y. (2017) · Signal Processing Image Communication · doi:10.1016/j.image.2017.01.005 · Cited by 21
Geometric and radiometric analysis of paintings
Brutto, M. L., Fabio Remondino, A., Brumana, Scaioni, Barazzetti, L., & Rizzi (2010) · Nova Science Publishers (Nova Science Publishers, Inc.) · Cited by 19
Heterogeneous domain adaptation network based on autoencoder
Wang, X., Ma, Y., Cheng, Y., Zou, L., & Rodrigues, J. J. P. C. (2017) · Journal of Parallel and Distributed Computing · doi:10.1016/j.jpdc.2017.06.003 · Cited by 20
Accuracy evaluation of structure from motion surface 3D reconstruction
Knyaz, V. A., & Zheltov, S. Y. (2017) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2272021 · Cited by 19
A Multiview Learning Framework With a Linear Computational Cost
Xue, X., Nie, F., Li, Z., Wang, S., Li, X., & Yao, M. (2017) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2017.2739423 · Cited by 22
Loose coupling visual-lidar odometry by combining VISO2 and LOAM
Yan, M., Wang, J., Li, J., & Zhang, C. (2017) · Chinese Control Conference (CCC) · doi:10.23919/chicc.2017.8028435 · Cited by 19
“As for the feature descriptors, we choose Sobel filter responses using the sum of absolute differences (SAD) error metric instead of SURF [8].”
MethodRadar emitter fingerprint recognition based on bispectrum and SURF feature
Kang, N., He, M., Han, J., et al. (2016) · CIE International Conference on Radar (RADAR) · doi:10.1109/radar.2016.8059588 · Cited by 19
“proposed, thus is also invariant in scale and rotation and is speeded up by integral image and Hessian matrix([6,10]).”
BackgroundMulti-modal and inertial sensor solutions for navigation-type factor graphs
Fourie, D. (2017) · Woods Hole Open Access Server (WHOAS) · doi:10.1575/1912/9305 · Cited by 18
“We had initially tried feature detection and descriptor matching BRISK [136], ORB [200], SURF [15] and SIFT [139], but found feature tracks to be relatively short at best.”
MethodAccelerated Embedded AKAZE Feature Detection Algorithm on FPGA
Kalms, L., Mohamed, K., & Göhringer, D. (2017) · International Symposium on Highly Efficient Accelerators and Reconfigurable Technologies (HEART) · doi:10.1145/3120895.3120898 · Cited by 19
“SIFT [13] and SURF [5] are the most widely used detectors in computer vision.”
MethodInfluentialUnsupervised learning from videos using temporal coherency deep networks
Redondo-Cabrera, C., & López-Sastre, R. J. (2018) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2018.08.003 · Cited by 18
Evaluation of Texture Descriptors for Validation of Counterfeit Documents
Berenguel, A., Terrades, O. R., Lladós, J., & Cañero, C. (2017) · IAPR International Conference on Document Analysis and Recognition (ICDAR) · doi:10.1109/icdar.2017.204 · Cited by 19
AMI-Net: Convolution Neural Networks With Affine Moment Invariants
Hao, Y., Li, Q., Mo, H., Zhang, H., & Li, H. (2018) · IEEE Signal Processing Letters · doi:10.1109/lsp.2018.2843296 · Cited by 19
Transformable Convolutional Neural Network for Text Classification
Xiao, L., Zhang, H., Chen, W., Wang, Y. K., & Jin, Y. (2018) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2018/625 · Cited by 18
“Traditionally, most works cope with above challenges by increasing the volume of datasets to cover more variants of features or employ the constant features [ Lowe, 1999; Bay et al. , 2006; Rublee et al. , 2011 ] .”
BackgroundFeature Selection for Unsupervised Domain Adaptation Using Optimal Transport
Gautheron, L., Redko, I., & Lartizien, C. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-10928-8_45 · Cited by 22
SweepLoc: Automatic Video-based Indoor Localization by Camera Sweeping
Li, M., Liu, N., Niu, Q., Liu, C., Chan, S.-H. G., & Gao, C. (2018) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3264930 · Cited by 20
“Apart from that, the employed local point descriptors (e.g. SURF [2], ORB [25]) are prone to blur and repetitive textures.”
MethodAn Experimental Comparison of ROS-compatible Stereo Visual SLAM Methods for Planetary Rovers
Giubilato, R., Chiodini, S., Pertile, M., & Debei, S. (2018) · IEEE International Workshop on Metrology for AeroSpace (MetroAeroSpace) · doi:10.1109/metroaerospace.2018.8453534 · Cited by 18
“Using feature detectors (SURF[14], SIFT[15], FAST[16] etc.”
Efficient Constellation-Based Map-Merging for Semantic SLAM
Frey, K. M., Steiner, T. J., & How, J. P. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794452 · Cited by 17
Li, Y., Jia, C., Wang, S., et al. (2018) · IEEE International Conference on Multimedia Big Data · doi:10.1109/bigmm.2018.8499170 · Cited by 19
“In the literatures, many feature coding schemes have been proposed to further reduce the bit-rate required for feature transmission, for both traditional features (e.g., SIFT [2], SURF [3], [4]) and deep features.”
BackgroundNo-Reference quality assessment of noisy images with local features and visual saliency models
Oszust, M. (2019) · Information Sciences · doi:10.1016/j.ins.2019.01.034 · Cited by 20
Energy-aware strategy for collaborative target-detection in wireless multimedia sensor network
Zam, A., Khayyambashi, M. R., & Bohlooli, A. (2019) · Multimedia Tools and Applications · doi:10.1007/s11042-019-7204-5 · Cited by 19
“State of the art features extraction algorithms such as histogram of orientation (HOG) [8], speeded up robust features (SURF) [4] or scale invariant feature transformation (SIFT) [20] have been developed essentially for single surveillance system or wired visual networks.”
MethodGarment Detectives: Discovering Clothes and Its Genre in Consumer Photos
Hidayati, S. C., Hua, K., Tsao, Y., et al. (2019) · IEEE Conference on Multimedia Information Processing and Retrieval (MIPR) · doi:10.1109/mipr.2019.00095 · Cited by 19
“Implementations: We used SURF [10] as the local feature detector and extracted the features at Canny edge points.”
MethodTS-NET: Combining Modality Specific and Common Features for Multimodal Patch Matching
En, S., Lechervy, A., & Jurie, F. (2018) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2018.8451804 · Cited by 27
“Pi-oneer works in patch matching [1] are based on handcrafted features such as the SIFT descriptor/detector or some variants, e.g. [2], DAISY, [3], etc.”
Accurate localization in dense urban area using Google street view images
Salarian, M., Manavella, A., & Ansari, R. (2015) · SAI Intelligent Systems Conference · doi:10.1109/intellisys.2015.7361184 · Cited by 19
An Automated Engineering Assistant: Learning Parsers for Technical Drawings
Daele, D. V., Decleyre, N., Dubois, H., & Meert, W. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i17.17783 · Cited by 36
Learning local representations for scalable RGB-D face recognition
Grati, N., Hamadou, A. B., & Hammami, M. (2020) · Expert Systems with Applications · doi:10.1016/j.eswa.2020.113319 · Cited by 19
Attention-Aware Age-Agnostic Visual Place Recognition
Wang, Z., Li, J., Khademi, S., & van Gemert, J. C. (2019) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2019.00181 · Cited by 28
“To extract good features for indexing, traditional works focus on handcrafted features such as SIFT[28]) and SURF[8].”
MethodA flexible and swift approach for 3D image–based survey in a cave
Alessandri, L., Baiocchi, V., Pizzo, S. D., et al. (2020) · Applied Geomatics · doi:10.1007/s12518-020-00309-4 · Cited by 19
“The followed procedure, fully automated in Metashape, is based on an automatic tie-point extractor algorithm, known as blob detector, which provides highly distinctive features invariant to image scaling and rotation, and partial invariant to brightness (Bay et al. 2006; Lowe 1999).”
BackgroundMethodHigh-resolution thermal face dataset for face and expression recognition
Kowalski, M., & Grudzień, A. (2018) · Metrology and Measurement Systems · doi:10.24425/119566 · Cited by 20
“Finally, global matching approaches such as Scale Invariant Feature Transform (SIFT) and Speeded-Up Robust Features (SURF) [28, 29] are used.”
MethodSTATE OF THE ART IN DIGITAL SURFACE MODELLING FROM MULTI-VIEW HIGH-RESOLUTION SATELLITE IMAGES
Han, Y., Wang, S., Gong, D., et al. (2020) · ISPRS annals of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-annals-v-2-2020-351-2020 · Cited by 19
Gaussian Process Gradient Maps for Loop-Closure Detection in Unstructured Planetary Environments
Gentil, C. L., Vayugundla, M., Giubilato, R., Stürzl, W., Vidal‐Calleja, T., & Triebel, R. (2020) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros45743.2020.9341667 · Cited by 19
“9 reports the precision obtained in all datasets as a function of the minimum number of inlier SURF matches after RANSAC n inl .”
BackgroundInfluentialRecent Trends on Object Detection and Image Classification: A Review
Athira, M. V., & Khan, D. M. (2020) · International Conference on Computational Performance Evaluation (ComPE) · doi:10.1109/compe49325.2020.9200080 · Cited by 19
“These architectures can be better understood in; [1] by Tzotzos et al. for SVM’s effectiveness on object based image analysis, [2] by Mangasarian et al. for working of SVM, [3] by Bay et al. for traditional detector and descriptor SURF (Speeded up Robust Features), [4] by Dalai et al. for HOG.”
Cheng, M., & Matsuoka, M. (2020) · International Journal of Remote Sensing · doi:10.1080/01431161.2020.1842538 · Cited by 18
Video shot boundary detection: A review
SenGupta, A., Thounaojam, D. M., Singh, K. M., et al. (2015) · International Conference on Electrical, Computer and Communication Technologies (ICECCT) · doi:10.1109/icecct.2015.7226084 · Cited by 19
Health information acquisition and position calculation of plug seedling in greenhouse seedling bed
Tong, J., Yu, J., Wu, C., Yu, G., Du, X., & Shi, H. (2021) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2021.106146 · Cited by 18
Image‐Based Indoor Localization Using Smartphone Camera
Li, S., Yu, B., Jin, Y., Huang, L. H., Zhang, H., & Liang, X. (2021) · Wireless Communications and Mobile Computing · doi:10.1155/2021/3279059 · Cited by 22
Tomită, M.-A., Zaffar, M., Milford, M. J., McDonald-Maier, K. D., & Ehsan, S. (2021) · IEEE Access · doi:10.1109/access.2021.3107778 · Cited by 21
Efficient Topology Estimation for Large Scale Optical Mapping
Elibol, A., Gracias, N., & García, R. (2012) · Springer tracts in advanced robotics · doi:10.1007/978-3-642-30313-5 · Cited by 21
Passive Copy Move Forgery Detection Using SURF, HOG and SIFT Features
Pandey, R. C., Agrawal, R., Singh, S. K., et al. (2014) · Advances in intelligent systems and computing · doi:10.1007/978-3-319-11933-5_74 · Cited by 19
Learning a Fast Emulator of a Binary Decision Process
Sochman, J., & Matas, J. (2007) · Asian Conference on Computer Vision · doi:10.1007/978-3-540-76390-1_24 · Cited by 17
SLAM in O(log n) with the Combined Kalman - Information filter
Cadena, C., & Neira, J. (2009) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2009.5354521 · Cited by 20
Badrinath, G. S., Gupta, P., & Mehrotra, H. (2011) · Journal of Real-Time Image Processing · doi:10.1007/s11554-011-0229-2 · Cited by 19
Shape context based object recognition and tracking in structured underwater environment
Han, K. M., & Choi, H. (2011) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss.2011.6049204 · Cited by 19
“Accordingly, distinct feature points based methods [19], [20], [21] are not suited to our needs despite these methods are popularly used in many other tracking algorithms.”
MethodLearning best views of 3D shapes from sketch contour
Zhao, L., Liang, S., Jia, J., et al. (2015) · The Visual Computer · doi:10.1007/s00371-015-1091-1 · Cited by 19
“Recent approaches [16,20] show that keypoint-based detectors,suchasdifferenceofGaussian(DoG)[19],Hessian operator [1] and Harris-Laplace detector [22], are suitable for line drawing feature extraction.”
BackgroundSelf-motion and wind velocity estimation for small-scale UAVs
Zachariah, D., & Jansson, M. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5979676 · Cited by 19
Monocular vision simultaneous localization and mapping using SURF
Zhang, Z., Huang, Y., Li, C., et al. (2008) · World Congress on Intelligent Control and Automation · doi:10.1109/wcica.2008.4593166 · Cited by 19
“Recently, an approximate scheme for SIFT called Speeded up Robust Feature (SURF) is provided by Bay et al [5].”
BackgroundMethodInfluential3D photon counting integral imaging with unknown sensor positions
Xiao, X., & Javidi, B. (2012) · Journal of the Optical Society of America A · doi:10.1364/josaa.29.000767 · Cited by 19
A fast algorithm for material image sequential stitching
Ma, B., Ban, X., Huang, H., Liu, W., Liu, C., Wu, D., & Zhi, Y. (2018) · Computational Materials Science · doi:10.1016/j.commatsci.2018.10.044 · Cited by 19
Features Combined Binary Descriptor Based on Voted Ring-Sampling Pattern
Liu, H., Zhang, Q., Fan, B., Wang, Z., & Han, J. (2019) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2943595 · Cited by 22
“To speed up the computational efficiency, Bay et al. [12] proposed SURF by using integral images, Haar wavelet and Hessian-matrix approximation.”
BackgroundMethodImage-Based Retrieval and Identification of Ancient Coins
Kampel, M., Huber-Mörk, R., & Zaharieva, M. (2009) · IEEE Intelligent Systems · doi:10.1109/mis.2009.29 · Cited by 19
High accuracy and language independent document retrieval with a Fast Invariant Transform
Liu, Q., Yano, H., Kimber, D., Liao, C., & Wilcox, L. D. (2009) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2009.5202515 · Cited by 19
“Unlike SIFT, SURF uses integral images to save keypoint finding and orientation assignment time.”
MethodInfluentialA Probabilistic Formulation of Keyword Spotting
Pérez, J. P. I. (2018) · RiuNet (Universitat Politècnica de València) · doi:10.4995/thesis/10251/116834 · Cited by 37
Kühne, J., Magno, M., & Benini, L. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3406948 · Cited by 33
“The feature descriptors, which are calculated on the dero-tated and rescaled feature points, are usually either composed of a binary gradient representation as in SIFT [23], SURF [24], or ORB [22], or composed of the actual derotated and rescaled image patch.”
Recent Advances in 3D Object Detection for Self-Driving Vehicles: A Survey
Fawole, O. A., & Rawat, D. B. (2024) · AI · doi:10.3390/ai5030061 · Cited by 31
“This technique in 3D object detection involves using feature extraction techniques, such as the Scale-Invariant Feature Transform (SIFT) [64] and the Speeded-Up Robust Features (SURF) [65].”
MethodUAV Localization Method with Keypoints on the Edges of Semantic Objects for Low-Altitude Economy
Li, Y., Zeng, Q., Shao, C., Zhuo, P., Li, B., & Sun, K. (2024) · Drones · doi:10.3390/drones9010014 · Cited by 19
Zhang, T., Dong, W., Li, Y., et al. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3662771 · Cited by 18
“Feature-based methods, such as scale-invariant feature transform (SIFT) [34] and speeded-up robust features (SURFs) [35], rely on detecting keypoints and matching them between images.”
How to localize humanoids with a single camera?
Alcantarilla, P. F., Stasse, O., Druon, S., Bergasa, L. M., & Dellaert, F. (2012) · Autonomous Robots · doi:10.1007/s10514-012-9312-1 · Cited by 20
Lossless compression of binary image descriptors for visual sensor networks
Ascenso, J., & Pereira, F. M. B. (2013) · IEEE International Conference on Digital Signal Processing (DSP) · doi:10.1109/icdsp.2013.6622692 · Cited by 18
Online multiple instance learning applied to hand detection in a humanoid robot
Ciliberto, C., Smeraldi, F., Natale, L., & Giorgio Metta, B. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6095002 · Cited by 18
An Idea to Apply Firefly Algorithm in 2D Image Key-Points Search
Woźniak, M., & Marszałek, Z. (2014) · Communications in computer and information science · doi:10.1007/978-3-319-11958-8_25 · Cited by 18
Appearance‐based landmark selection for visual localization
Bürki, M., Cadena, C., Gilitschenski, I., Siegwart, R., & Nieto, J. I. (2019) · Journal of Field Robotics · doi:10.1002/rob.21870 · Cited by 18
“Lategahn, Beck, and Stiller (2014) propose a local feature descriptor named DIRD which exhibits illumination invariance superior to other popular local features such as SURF (Bay, Tuytelaars, & Van Gool, 2006) or BRIEF (Calonder et al., 2012).”
BackgroundTree bark re-identification using a deep-learning feature descriptor
Robert, M., Dallaire, P., & Giguère, P. (2020) · Canadian Conference on Computer and Robot Vision · doi:10.1109/crv50864.2020.00012 · Cited by 22
Real-time defect detection network for polarizer based on deep learning
Liu, R., Sun, Z., Wang, A., Yang, K., Wang, Y., & Sun, Q. (2020) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-020-01536-7 · Cited by 20
“The commonly used feature representation algorithms include scale-invariant feature transforms (SIFT) (Lowe 2004), histogram of oriented gradient (HOG) (Dalal and Triggs 2005), local binary pattern (LBP) (Ahonen et al. 2006), and speeded up robust features (SURF) (Bay et al. 2006).”
MethodNgọc, V. T. N., Agwu, A. C., Sơn, L. H., Tuấn, T. M., Giap, C. N., Thanh, M. T. G., Duy, H. B., & Ngân, T. T. (2020) · Diagnostics · doi:10.3390/diagnostics10040209 · Cited by 19
Real-time Bangladeshi Currency Detection System for Visually Impaired Person
Sarker, M. F. R., Raju, M. I. M., Marouf, A. A., Hafiz, R., Hossain, S. A., & Protik, M. H. K. (2019) · International Conference on Bangla Speech and Language Processing (ICBSLP) · doi:10.1109/icbslp47725.2019.201518 · Cited by 19
“Template matching [6], shape descriptors [7] etc. are outperformed by regular feature descriptors such as scale-invariant feature transform (SIFT) [8], speeded-up robust feature (SURF) [9], maximally stable external regions (MSER) [10] etc.”
Background
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.