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 901–1,000 of 14,752
Toward real-time grocery detection for the visually impaired
Winlock, T., Christiansen, E., & Belongie, S. (2010) · 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops · doi:10.1109/cvprw.2010.5543576 · Cited by 77
Reversible Data Hiding in Color Image With Grayscale Invariance
Hou, D., Zhang, W., Chen, K., et al. (2018) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2018.2803303 · Cited by 77
“As far as we know, almost all of SIFT or SIFT-extended descriptors like SURF (Speeded Up Robust Features) [37] and their applications only operate on the luminance channel.”
MethodThe Smartphone As a Platform for Wearable Cameras in Health Research
Gurrin, C. G., Qiu, Z., Hughes, M. M., Caprani, N., Doherty, A., Hodges, S. E., & Smeaton, A. F. (2013) · American Journal of Preventive Medicine · doi:10.1016/j.amepre.2012.11.010 · Cited by 91
PhotoNet: A Similarity-Aware Picture Delivery Service for Situation Awareness
Uddin, M. Y. S., Wang, H., Saremi, F., Qi, G.-J., Abdelzaher, T., & Huang, T. S. (2011) · IEEE Real-Time Systems Symposium (RTSS) · doi:10.1109/rtss.2011.36 · Cited by 76
3D Scene Reconstruction from Multiple Spherical Stereo Pairs
Kim, H., & Hilton, A. (2013) · International Journal of Computer Vision · doi:10.1007/s11263-013-0616-1 · Cited by 73
Flowing on Riemannian Manifold: Domain Adaptation by Shifting Covariance
Cui, Z., Li, W., Xu, D., Shan, S., Chen, X., & Li, X. (2014) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2014.2305701 · Cited by 69
“Similar to [1], [2], and [5], we extract SURF features [33] and encode the images with 800-bin token frequency (TF) feature by using a codebook trained from a subset of Amazon images.”
MethodTemporal Surface Tracking Using Mesh Evolution
Varanasi, K., Zaharescu, A., Boyer, E., & Horaud, R. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-88688-4_3 · Cited by 81
Mehmood, Z., Abbas, F. M., Mahmood, T., Javid, M. A., Rehman, A. U., & Nawaz, T. (2018) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-018-3062-0 · Cited by 81
“In recent proposed CBIR techniques, various feature descriptors like scale-invariant feature transform (SIFT) [5], SURF [6], HOG [7], binary robust invariants scalable key-points (BRISK) [8], and maximally stable extremal regions (MSER) [9] were being used to improve the performance of image…”
BackgroundMethodInfluentialStructure-Guided Image Inpainting Using Homography Transformation
Liu, J., Yang, S., Fang, Y., et al. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2018.2831636 · Cited by 76
“We begin with speeded up robust features (SURF, [35]) point detection and compute SURF descriptors for each feature point.”
MethodA 3D-CNN and LSTM Based Multi-Task Learning Architecture for Action Recognition
Ouyang, X., Xu, S., Zhang, C., Zhou, P., Yang, Y., Liu, G., & Li, X. (2019) · IEEE Access · doi:10.1109/access.2019.2906654 · Cited by 82
“trations of such successful features are SIFT [37], SURF [38], ORB [39].”
MethodA Joint Optimization Approach of LiDAR-Camera Fusion for Accurate Dense 3-D Reconstructions
Zhen, W., Hu, Y., Liu, J., & Scherer, S. (2019) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2019.2928261 · Cited by 73
Dual-Resolution Correspondence NetworksPreprint
Li, X., Han, K., Li, S., & Prisacariu, V. A. (2020) · Oxford University Research Archive (ORA) (University of Oxford) · doi:10.48550/arxiv.2006.08844 · Cited by 128
Zhang, A., Lipton, Z. C., Mu, L., & Smola, A. J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.11342 · Cited by 76
“SIFT (Lowe, 2004), SURF (Bay et al., 2006), HOG (histograms of oriented gradient) (Dalal & Triggs, 2005), bags of visual words90 and similar feature extractors ruled the roost.”
MethodSuper Fast Event Recognition in Internet Videos
Jiang, Y., Dai, Q., Mei, T., Rui, Y., & Chang, S. (2015) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2015.2436813 · Cited by 74
Fusing Time-of-Flight Depth and Color for Real-Time Segmentation and Tracking
Bleiweiss, A., & Werman, M. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-03778-8_5 · Cited by 75
An efficient ear recognition technique invariant to illumination and pose
Prakash, S., & Gupta, P. (2011) · Telecommunication Systems · doi:10.1007/s11235-011-9621-2 · Cited by 76
He, W. P., Yamashita, T., Lu, H., & Lao, S. (2009) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2009.5459360 · Cited by 93
“Recently developed SURF feature [2] is a variant of SIFT and shares equal repeatability, distinctiveness and robustness, but has much faster computing speed.”
BackgroundYousuf, M., Mehmood, Z., Habib, H. A., et al. (2018) · Mathematical Problems in Engineering · doi:10.1155/2018/2134395 · Cited by 75
Assigning Visual Words to Places for Loop Closure Detection
Tsintotas, K. A., Bampis, L., & Gasteratos, A. C. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8461146 · Cited by 77
“At query time, SURF [5] local feature descriptors are converted into their nearest neighboring VWs, while the votes’ aggregation (density of dashed arrows) indicates an image-to-place match.”
BackgroundMethodInfluentialImage-Based Geo-Localization Using Satellite Imagery
Hu, S., & Lee, G. H. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01186-0 · Cited by 91
“This causes cross-view matching with traditional handcrafted features like SIFT (Lowe 2004) and SURF (Bay et al. 2006) fail.”
MethodDUT: Learning Video Stabilization by Simply Watching Unstable Videos
Xu, Y., Zhang, J., Maybank, S. J., & Tao, D. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3182887 · Cited by 93
R-PointHop: A Green, Accurate, and Unsupervised Point Cloud Registration Method
Kadam, P., Zhang, M., Liu, S., & Kuo, C. ‐. J. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3160609 · Cited by 89
Joint Hand Motion and Interaction Hotspots Prediction from Egocentric Videos
Liu, S., Tripathi, S., Majumdar, S., & Wang, X. (2022) · 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/cvpr52688.2022.00328 · Cited by 155
Ozog, P., Carlevaris‐Bianco, N., Kim, A., & Eustice, R. M. (2015) · Journal of Field Robotics · doi:10.1002/rob.21582 · Cited by 82
Efficient Non-Consecutive Feature Tracking for Robust Structure-From-Motion
Zhang, G., Liu, H., Dong, Z., Jia, J., Wong, T., & Bao, H. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2607425 · Cited by 77
Experiments in place recognition using gist panoramas
Murillo, A. C., & Košecká, J. (2009) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2009.5457552 · Cited by 78
“using [9, 1], between the query panorama properly aligned to the reference matched panoramas.”
MethodThermal–visible registration of human silhouettes: A similarity measure performance evaluation
Bilodeau, G.-A., Torabi, A., St-Charles, P.-L., & Riahi, D. (2014) · Infrared Physics & Technology · doi:10.1016/j.infrared.2014.02.005 · Cited by 74
Multilevel SIFT Matching for Large-Size VHR Image Registration
Huo, C., Pan, C., Huo, L., & Zhou, Z. (2011) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2011.2163491 · Cited by 76
“[4] bring new potentials for feature-based remote sensing image registration due to the scale invariance of the detector and the distinctiveness of the descriptor.”
BackgroundVousdoukas, M. I., Ferreira, P., Almeida, L. P., et al. (2011) · Ocean Dynamics · doi:10.1007/s10236-011-0440-5 · Cited by 75
Collecting and Analyzing Eye-Tracking Data in Outdoor Environments
Evans, K. M., Jacobs, R. A., Tarduno, J. A., & Pelz, J. B. (2012) · Journal of Eye Movement Research · doi:10.16910/jemr.5.2.6 · Cited by 99
Using grayscale images for object recognition with convolutional-recursive neural network
Bui, H. M., Lech, M., Cheng, E., Neville, K. L., & Burnett, L. (2016) · International Conference on Communications and Electronics (ICCE) · doi:10.1109/cce.2016.7562656 · Cited by 98
“In the early days of object recognition, grayscale images were used quite successfully with manually selected features such as HoG [2], SIFT [4] and SURF [6].”
MethodMahmood, T., Irtaza, A., Mehmood, Z., & Mahmood, M. T. (2017) · Forensic Science International · doi:10.1016/j.forsciint.2017.07.037 · Cited by 74
Fast copy-move forgery detection using local bidirectional coherency error refinement
Bi, X., & Pun, C. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2018.03.028 · Cited by 79
CNN‐based reference comparison method for classifying bare PCB defects
Peng, W., Liu, C., Liu, M., Gao, Y., & Liu, H. (2018) · The Journal of Engineering · doi:10.1049/joe.2018.8271 · Cited by 79
Omnidirectional CNN for Visual Place Recognition and Navigation
Wang, T.-H. J., Huang, H.-J., Lin, J.-T., Hu, C.-W., Zeng, K.-H., & Sun, M. (2018) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2018.8463173 · Cited by 92
“Such methods include SIFT [28], SURF [9], ORB [39], which was recently used in ORB-SLAM system [33], etc.”
MethodOne-Shot-Learning Gesture Recognition Using HOG-HOF Features
Konečný, J., & Hagara, M. (2017) · The Springer series on challenges in machine learning · doi:10.1007/978-3-319-57021-1_12 · Cited by 74
On-Device Mobile Visual Location Recognition by Integrating Vision and Inertial Sensors
Guan, T., He, Y., Gao, J., Yang, J., & Yu, J. (2013) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2013.2265674 · Cited by 74
“For our Wuhan database, both the BOF and gravity-aligned VLAD descriptors are generated by using the gravity-aligned SURF features.”
MethodInfluentialA Method for Counting People in Crowded Scenes
Conte, D., Foggia, P., Percannella, G., Tufano, F., & Vento, M. (2010) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2010.78 · Cited by 77
Generic 3D Representation via Pose Estimation and Matching
Zamir, A. R., Wekel, T., Agrawal, P., Wei, C., Malik, J., & Savarese, S. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46487-9_33 · Cited by 69
“Therefore, we presume (and empirically observe) that the generality of our 3D representation is mostly attributed to the camera pose estimation component. scribing, and matching image features, either through a handcrafting the feature [35,11,39,38,5,40] or learning it [50,25,65,9,66,49,19].”
BackgroundOn the generalization of color texture-based face anti-spoofing
Boulkenafet, Z., Komulainen, J., & Hadid, A. (2018) · Image and Vision Computing · doi:10.1016/j.imavis.2018.04.007 · Cited by 78
Moving Indoor: Unsupervised Video Depth Learning in Challenging Environments
Zhou, J., Wang, Y., Qin, K., & Zeng, W. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00871 · Cited by 83
“The sparse matchings are also obtained by SURF [1].”
MethodA Survey of Vehicle Re-Identification Based on Deep Learning
Wang, H., Hou, J., & Chen, N. (2019) · IEEE Access · doi:10.1109/access.2019.2956172 · Cited by 77
“shape context [24], spin image [25], Speeded Up Robust Features [26] (SURF), Space-Time Interest Points [27] (STIP), Histogram of ori-”
BackgroundMulti-scale local cues and hierarchical attention-based LSTM for stock price trend prediction
Teng, X., Zhang, X., & Luo, Z. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.07.016 · Cited by 77
Lynen, S., Bosse, M., Furgale, P., & Siegwart, R. (2014) · 2014 2nd International Conference on 3D Vision · doi:10.1109/3dv.2014.36 · Cited by 73
The SmartVision local navigation aid for blind and visually impaired persons
-, J. J., Farrajota, M., Rodrigues, J. M. F., & du Buf, J. M. H. (2011) · International Journal of Digital Content Technology and its Applications · doi:10.4156/jdcta.vol5.issue5.40 · Cited by 89
Automatic detection of child pornography using color visual words
Ulges, A., & Stahl, A. (2011) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2011.6011977 · Cited by 85
Colour Mapping: A Review of Recent Methods, Extensions and Applications
Faridul, H. S., Pouli, T., Chamaret, C., Stauder, J., Reinhard, E., Kuzovkin, D., & Trémeau, A. (2015) · Computer Graphics Forum · doi:10.1111/cgf.12671 · Cited by 85
“These include often-used algorithms such as scale-invariant feature transform (SIFT) [Low04] and speeded-up robust features (SURFs) [BTVG06].”
MethodRobust Real-Time Pattern Matching Using Bayesian Sequential Hypothesis Testing
Pele, O., & Werman, M. (2008) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2007.70794 · Cited by 77
Handling Urban Location Recognition as a 2D Homothetic Problem
Baatz, G., Köser, K., Chen, D. M., Grzeszczuk, R., & Pollefeys, M. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-15567-3_20 · Cited by 70
Nearest neighbour classification of Indian sign language gestures using kinect camera
Ansari, Z., & Harit, G. (2016) · Sadhana · doi:10.1007/s12046-015-0405-3 · Cited by 80
“The descriptor ‘describes the distribution of the intensity content within the interest point neighbourhood, similar to the gradient information extracted by SIFT and its variants [48].’”
MethodCelik, C., & Bılge, H. Ş. (2017) · Pattern Recognition · doi:10.1016/j.patcog.2017.03.006 · Cited by 78
Methods and Challenges in Shot Boundary Detection: A Review
Abdulhussain, S. H., Ramli, A. R., Saripan, M. I., Mahmmod, B. M., Al-Haddad, S. A. R., & Jassim, W. A. (2018) · Entropy · doi:10.3390/e20040214 · Cited by 84
A Microforce-Sensing Mobile Microrobot for Automated Micromanipulation Tasks
Jing, W., Chowdhury, S., Guix, M., Wang, J.-X., An, Z., Johnson, B., & Cappelleri, D. J. (2018) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2018.2833810 · Cited by 74
Retina Mosaicing Using Local Features
Cattin, P. C., Bay, H., Van Gool, L., & Székely, G. (2006) · Lecture Notes in Computer Science · doi:10.1007/11866763_23 · Cited by 75
“The use of our very recently developed framework [11] proved more successful.”
BackgroundMethodInfluentialSelf-citationLearning automatic concept detectors from online video
Ulges, A., Schulze, C., Koch, M., & Breuel, T. M. (2009) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2009.08.002 · Cited by 72
Automatic visual inspection of printed circuit board for defect detection and classification
Chaudhary, V., Dave, I. R., & Upla, K. P. (2017) · 2017 International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET) · doi:10.1109/wispnet.2017.8299858 · Cited by 59
Pose Graph optimization for Unsupervised Monocular Visual Odometry
Li, Y., Ushiku, Y., & Harada, T. (2019) · 2019 International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra.2019.8793706 · Cited by 71
“Feature-based approach usually consists of two stages: 1) data association, through hand-engineered feature extraction (e.g. SURF [19], SIFT [20] and ORB [21]) and matching, and 2) pose estimation via minimizing feature point re-projection error.”
MethodA low-cost UAV framework towards ornamental plant detection and counting in the wild
Bayraktar, E., Basarkan, M. E., & Çelebi, N. (2020) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2020.06.012 · Cited by 76
Zhang, Z., Zhang, L., Wu, J., & Guo, W. (2024) · IEEE Geoscience and Remote Sensing Magazine · doi:10.1109/mgrs.2024.3404506 · Cited by 110
VIRaL: Visual Image Retrieval and Localization
Kalantidis, Y., Tolias, G., Avrithis, Y., Phiniketos, M., Spyrou, E., Mylonas, P., & Kollias, S. (2010) · Multimedia Tools and Applications · doi:10.1007/s11042-010-0651-7 · Cited by 77
Improving Harris corner selection strategy
Bellavia, F., Tegolo, D., & Valenti, C. F. (2011) · IET Computer Vision · doi:10.1049/iet-cvi.2009.0127 · Cited by 72
Real-Time Model-Based Video Stabilization for Microaerial Vehicles
Aguilar, W. G., & Ángulo, C. (2015) · Neural Processing Letters · doi:10.1007/s11063-015-9439-0 · Cited by 77
“A list of algorithms performing this features description task can be found in the literature [12,21,32], but binary robust invariant scalable keypoints [27], fast retina keypoint [8], oriented FAST and rotated BRIEF [34], scale invariant feature transform [29] and speed up robust feature (SURF) [11,30] are common in…”
MethodMultiview Classification With Cohesion and Diversity
Tao, H., Hou, C., Yi, D., & Zhu, J. (2018) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2018.2881474 · Cited by 75
“We extract six visual features: GIST, LBP, PHOG, SIFT, SURF [40], and wavelet texture (WT).”
MethodAdaptive Channel Selection for Robust Visual Object Tracking with Discriminative Correlation Filters
Xu, T., Feng, Z., Wu, X., & Kittler, J. (2021) · International Journal of Computer Vision · doi:10.1007/s11263-021-01435-1 · Cited by 79
“Typical examples of ROI-based features include Scale-InvariantFeatureTransform(SIFT)(Lowe1999)andSpeeded Up Robust Features (SURF) (Bay et al. 2006).”
MethodSSSIC: Semantics-to-Signal Scalable Image Coding With Learned Structural Representations
Yan, N., Gao, C., Liu, D., Li, H., Li, L., & Wu, F. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3121131 · Cited by 87
Contactless palmprint and knuckle biometrics for mobile devices
Choraś, M., & Kozik, R. (2011) · Pattern Analysis and Applications · doi:10.1007/s10044-011-0248-4 · Cited by 72
Location-based augmented reality on mobile phones
Paucher, R., & Turk, M. (2010) · 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2010.5543249 · Cited by 76
BoRF: Loop-closure detection with scale invariant visual features
Zhang, H. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5980273 · Cited by 70
“In this work, we assume the use of a scale-invariant feature (e.g., SIFT or SURF), which has emerged as the leading approach to feature extraction in a variety of applications including robot navigation.”
MethodInfluentialOn rotation invariance in copy-move forgery detection
Christlein, V., Rieß, C., & Angelopoulou, E. (2010) · IEEE International Workshop on Information Forensics and Security · doi:10.1109/wifs.2010.5711472 · Cited by 74
Adeli, H., Vitu, F., & Zelinsky, G. J. (2016) · Journal of Neuroscience · doi:10.1523/jneurosci.0825-16.2016 · Cited by 86
Simple yet stable bearing‐only navigation
Krajník, T., Faigl, J., Vonásek, V., Košnar, K., Kulich, M., & Přeučil, L. (2010) · Journal of Field Robotics · doi:10.1002/rob.20354 · Cited by 85
The structure-from-motion reconstruction pipeline – a survey with focus on short image sequences
Häming, K., & Peters, G. (2010) · Czech digital mathematics library · Cited by 101
Edge-SIFT: Discriminative Binary Descriptor for Scalable Partial-Duplicate Mobile Search
Zhang, S. S., Tian, Q., Lü, K., Huang, Q., & Gao, W. (2013) · IEEE Transactions on Image Processing · doi:10.1109/tip.2013.2251650 · Cited by 73
Visual Homing From Scale With an Uncalibrated Omnidirectional Camera
Liu, M., Pradalier, C., & Siegwart, R. (2013) · IEEE Transactions on Robotics · doi:10.1109/tro.2013.2272251 · Cited by 75
Flight Security and Safety of Drones in Airborne Fog Computing Systems
He, D., Qiao, Y., Chan, S., & Guizani, N. (2018) · IEEE Communications Magazine · doi:10.1109/mcom.2018.1700916 · Cited by 70
“[14] H.”
MethodBakour, K., & Ünver, H. M. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-05816-y · Cited by 74
“SURF (Speeded Up Robust Features) [31] This feature has been introduced as an alternative for the SIFT algo-rithm.”
MethodKeypoint Signatures for Fast Learning and Recognition
Calonder, M., Lepetit, V., & Fua, P. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-88682-2_6 · Cited by 78
Deep shot: a framework for migrating tasks across devices using mobile phone cameras
Chang, T.-H., & Li, Y. (2011) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/1978942.1979257 · Cited by 72
“It then extracts visual features from the screenshot and the picture taken by the camera, using a computer vision algorithm, Speeded-Up Robust Features (SURF) [4].”
MethodAn interactive augmented reality coloring book
Clark, A., & Duenser, A. (2012) · IEEE Symposium on 3D User Interfaces (3DUI) · doi:10.1109/3dui.2012.6184168 · Cited by 111
“Natural feature registration involves searching for features which are present in an existing image, and describing them in a way which ensures they are identifiable and unique regardless of changes in viewpoint, orientation, scale, and other transformations [12, 13].”
Illumination-invariant image matching for autonomous UAV localisation based on optical sensing
Wan, X., Liu, J. G., Yan, H., & Morgan, G. L. K. (2016) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2016.05.016 · Cited by 80
Liu, M., Han, S., & Lee, S. H. (2016) · Construction Innovation · doi:10.1108/ci-10-2015-0054 · Cited by 71
3D point cloud map based vehicle localization using stereo camera
Xu, Y., John, V., Mita, S., Tehrani, H., Ishimaru, K., & Nishino, S. (2017) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2017.7995765 · Cited by 85
Local Metric Learning for Exemplar-Based Object Detection
You, X., Li, Q., Tao, D., Ou, W., & Gong, M. (2014) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2014.2306031 · Cited by 70
“Simultaneously, those informative histogram features such as SIFT [3], SURF [4], LBP [5], and HOG [6], have also been widely used to reliably represent objects.”
MethodSalient Feature Region: A New Method for Retinal Image Registration
Zheng, J., Tian, J., Deng, K., Dai, X., Zhang, X., & Xu, M. (2010) · IEEE Transactions on Information Technology in Biomedicine · doi:10.1109/titb.2010.2091145 · Cited by 69
“1) Local Feature Descriptor: Local feature descriptor can either be content based, such as SIFT, PCA-SIFT [22], and SURF [23] or local relationship based that contains geometric information and topological information [8].”
BackgroundFace recognition using adaptive local ternary patterns method
Yang, W., Wang, Z., & Zhang, B. (2016) · Neurocomputing · doi:10.1016/j.neucom.2015.11.134 · Cited by 73
RGB-T SLAM: A flexible SLAM framework by combining appearance and thermal information
Chen, L. F., Sun, L., Yang, T., Fan, L., Huang, K., & Xuanyuan, Z. (2017) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2017.7989668 · Cited by 82
“During tracking, they derive Lucas-Kanade [14] optical flow method, but when the thermal-infrared camera corrects the non-uniformity and an apparent delay incurs, SURF [15] features are extracted between recent consecutive frames instead.”
MethodDeep convolutional features for image retrieval
Gkelios, S., Sophokleous, A., Plakias, S., Boutalis, Y. S., & Chatzichristofis, S. A. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.114940 · Cited by 82
Deep learning serves traffic safety analysis: A forward‐looking review
Razi, A., Chen, X., Li, H., Wang, H., Russo, B. J., Chen, Y., & Yu, H. (2022) · IET Intelligent Transport Systems · doi:10.1049/itr2.12257 · Cited by 98
Online Spatio-temporal Structural Context Learning for Visual Tracking
Wen, L., Cai, Z., Lei, Z., Yi, D., & Li, S. Z. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33765-9_51 · Cited by 76
Extracting general-purpose features from LIDAR data
Li, Y., & Olson, E. (2010) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/robot.2010.5509690 · Cited by 77
Redondi, A. E. C., Baroffio, L., Cesana, M., et al. (2013) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp.2013.6659301 · Cited by 70
“Visual features extraction may be performed through state-of-the-art algorithms [12][13][14], or by selecting the best features for a particular analysis task through supervised methods [15] or information theory [16].”
MethodOnline and Incremental Appearance-based SLAM in Highly Dynamic Environments
Kawewong, A., Tongprasit, N., Tangruamsub, S., & Hasegawa, O. (2011) · The International Journal of Robotics Research · doi:10.1177/0278364910371855 · Cited by 70
Fast geometric point labeling using conditional random fields
Rusu, R. B., Holzbach, A., Blodow, N., et al. (2009) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2009.5354763 · Cited by 70
Accuracy of Reconstruction of the Tree Stem Surface Using Terrestrial Close-Range Photogrammetry
Surový, P., Yoshimoto, A., & Panagiotidis, D. (2016) · Remote Sensing · doi:10.3390/rs8020123 · Cited by 70
“In another study, [18] used a novel approach for acquisition of forest horizontal structure (stem distribution) with panoramic photography or a similar technique which may be applied through smartphone [19]; however, a recent technique using multi-view stereopsis (MVS) combined with computer vision and photogrammetry,…”
MethodAugmented vision and interactive monitoring in 3D printing process
Ceruti, A., Liverani, A., & Bombardi, T. (2016) · International Journal on Interactive Design and Manufacturing (IJIDeM) · doi:10.1007/s12008-016-0347-y · Cited by 72
“According to the paper by Bay et al. [17], once defined the position p = [ x , y ] of a pixel in an image I , the Hessian matrix H ( p , σ) in p at scale σ (greater the value of σ , greater the blur) can be defined as: The L xx ( p , σ) operator represents the convolution of the Gaussian second order…”
MethodLoop closure detection for visual SLAM systems using convolutional neural network
Zhang, X., Su, Y. L., & Zhu, X. (2017) · International Conference on Automation and Computing (ICAC) · doi:10.23919/iconac.2017.8082072 · Cited by 79
“We compare the proposed CNN-based visual loop closure detection method against stateof-the-art methods FAB-MAP [11] (which is based on handcrafted features: SURF [16]) on several different datasets using precision-recall curve metric.”
MethodImproving optimization of convolutional neural networks through parameter fine-tuning
Becherer, N., Pecarina, J. M., Nykl, S. L., & Hopkinson, K. M. (2017) · Neural Computing and Applications · doi:10.1007/s00521-017-3285-0 · Cited by 75
Mapping, navigation, and learning for off‐road traversal
Konolige, K., Agrawal, M., Blas, M. R., Bolles, R. C., Gerkey, B. P., Solà, J., & Sundaresan, A. (2008) · Journal of Field Robotics · doi:10.1002/rob.20271 · Cited by 74
In the wild image retrieval and clustering for 3D cultural heritage landmarks reconstruction
Makantasis, K., Doulamis, A. D., Doulamis, N. D., & Ioannides, M. (2014) · Multimedia Tools and Applications · doi:10.1007/s11042-014-2191-z · Cited by 76
Informative feature selection for object recognition via Sparse PCA
Naikal, N., Yang, A. Y., & Sastry, S. (2011) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2011.6126321 · Cited by 65
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.