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 7,301–7,400 of 14,752
Multiple Kernel Learning and Optical Flow for Action Recognition in RGB-D Video
Viet, V. H., Ly, N. Q., Sơn, T. T., & Hoang, P. M. (2015) · International Conference on Knowledge and Systems Engineering/Proceedings · doi:10.1109/kse.2015.39 · Cited by 5
“Beside global feature approaches, the local features methods such as: histogram of 3D oriented gradients (HOG3D) [1], histogram of optical flow (HOF) [9, 10], 3D speeded up robust features (SURF3D) [8] extends from SURF [4], 3D scale invariant feature transforms (3D-SIFT) [22] extends from SIFT [6] and dense trajector…”
MethodExamination of Hybrid Image Feature Trackers
Abeles, P. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-41939-3_54 · Cited by 6
“SURF is a popular state-of-the-art descriptor designed for speed and stability.”
BackgroundInfluentialTrifocal system for high-quality inter-camera mapping and virtual view synthesis
Seitner, F. H., Nezveda, M., Gelautz, M., et al. (2015) · International Conference on D Imaging · doi:10.1109/ic3d.2015.7391819 · Cited by 5
“In particular, we find putative point correspondences in between the views using SURF [3] and perform an uncalibrated stereo rectification using [23].”
MethodObservability Criteria and Null-Measurement Kalman Filter for Vision-Aided Inertial Navigation
Schmitt, L., & Fichter, W. (2016) · Journal of Guidance Control and Dynamics · doi:10.2514/1.g001146 · Cited by 5
Yamazaki, K. (2015) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio.2015.7418919 · Cited by 8
“Although well-known image features [10] [13] [16] provide highly reliable results for the purpose of object detection, these features cannot be used for clothing because there is an assumption that the transformations are limited only to solid objects.”
BackgroundVisual location recognition using smartphone sensors for indoor environment
Feng, G., Tan, X., & Ma, L. (2015) · International Conference for Internet Technology and Secured Transactions · doi:10.1109/icitst.2015.7412135 · Cited by 5
Advanced tools and framework for historical film restoration
Croci, S., Aydın, T. O., Stefanoski, N., Groß, M. H., & Smolić, A. (2017) · Journal of Electronic Imaging · doi:10.1117/1.jei.26.1.011021 · Cited by 8
Serej, N. D., Ahmadian, A., Kasaei, S., et al. (2015) · Signal Image and Video Processing · doi:10.1007/s11760-015-0849-2 · Cited by 5
Kawasaki, A., Saitô, H., & Hara, K. (2015) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2015.7350996 · Cited by 5
“Since the viewing directions of those cameras are very different, a conventional algorithm such as SURF[8] can not find enough correspondences.”
MethodInfluentialSocio-mobile landmark recognition using local features with adaptive region selection
Zhang, C., Zhang, Y., Zhu, X., Xue, Z., Qin, L., Huang, Q., & Tian, Q. (2015) · Neurocomputing · doi:10.1016/j.neucom.2014.10.105 · Cited by 5
Speed-up Feature Detector using Adaptive Accelerated Segment Test
Biadgie, Y., & Sohn, K.-A. (2016) · IETE Technical Review · doi:10.1080/02564602.2015.1103669 · Cited by 6
“In SIFT [3,4] and SURF (speeded up robust features) [31], each octave of the pyramid has six intraoctaves.”
BackgroundMethodInfluentialUnsupervised Indoor Scene Classification Based on Context for a Mobile Robot
Madokoro, H., Utsumi, Y., & Sato, K. (2013) · Journal of the Robotics Society of Japan · doi:10.7210/jrsj.31.918 · Cited by 5
Matúška, S., Hudec, R., Kamencay, P., et al. (2016) · Radioengineering · doi:10.13164/re.2016.0161 · Cited by 5
TokyoTech+Canon at TRECVID 2011
Inoue, N., Mori, K., Liang, Z., et al. (2011) · Cited by 5
Blending 2D and 3D Face Recognition
Tistarelli, M., Cadoni, M. I., Lagorio, A., & Grosso, E. (2016) · Springer eBooks · doi:10.1007/978-3-319-28501-6_13 · Cited by 6
RECOD at MediaEval 2015: Affective Impact of Movies Task
Moreira, D., Avila, S., Pérez, M., et al. (2015) · MediaEval · Cited by 5
Mixed Reality for Improving Tele-rehabilitation Practices
Gervasi, O., Magni, R., & Riganelli, M. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-21404-7_42 · Cited by 5
Hybrid Feature and Template Based Tracking for Augmented Reality Application
Negara, I. G. P. K., Teck, F. W., & Li, Y. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16634-6_28 · Cited by 5
Collaborative Mobile 3D Reconstruction of Urban Scenes
Tanács, A., Majdik, A. L., Hajder, L., Molnár, J., Sánta, Z., & Kató, Z. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16634-6_36 · Cited by 5
Vision based obstacle avoidance and recognition system
Verma, N. K., Kumar, G., Siddhant, A., Nama, P., Raj, A., Mustafa, A., Dhar, N. K., & Salour, A. (2015) · IEEE Workshop on Computational Intelligence: Theories, Applications and Future Directions (WCI) · doi:10.1109/wci.2015.7495534 · Cited by 7
Exploring the DolphinSLAM's parameters
Zaffari, G. B., dos Santos, M. M., Duarte, A. C., et al. (2016) · OCEANS 2016 - Shanghai · doi:10.1109/oceansap.2016.7485531 · Cited by 5
“In the DolphinSLAM algorithm it is possible to choose between the Scale Invariant Feature Transform (SIFT) method [13] and the Speeded Up Robust Features (SURF) method [14] for image feature extraction.”
Resolution adaptive feature extracting and fusing framework for person re-identification
Yang, H., Wang, X., Zhu, J., et al. (2016) · Neurocomputing · doi:10.1016/j.neucom.2016.03.096 · Cited by 5
Bag-of-visual-words model for artificial pornographic images recognition
Li, F., Luo, S., Liu, X., & Zou, B. (2016) · Journal of Central South University · doi:10.1007/s11771-016-3190-1 · Cited by 5
“Local feature points in sex scene were firstly detected [22] using the SURF [23] descriptors, and then high-level semantic dictionary was constructed.”
BackgroundMethodPuybareau, É., Talbot, H., Béquignon, É., et al. (2016) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2016.7532556 · Cited by 5
ROS-based ground stereo vision detection: implementation and experiments
Hu, T., Zhao, B., Tang, D., et al. (2016) · Robotics and Biomimetics · doi:10.1186/s40638-016-0046-y · Cited by 5
“SIFT [14], SURF [15], ORB [16], FAST [17] and BRISK [18] corner detectors are, respectively, tested with the dataset.”
MethodKey point descriptor based copy and move image forgery detection system
Nithiya, R., & Veluchamy, S. (2016) · International Conference on Science Technology Engineering and Mathematics (ICONSTEM) · doi:10.1109/iconstem.2016.7560959 · Cited by 5
Model based Detection and 3D Localization of Planar Objects for Industrial Setups
Sakçak, B., Bascetta, L., & Ferretti, G. (2016) · International Conference on Informatics in Control, Automation and Robotics (ICINCO) · doi:10.5220/0005982503600367 · Cited by 5
“For that purpose one popular approach is to use invariant feature descriptors such as SIFT (Lowe, 2004), SURF (Bay et al., 2006), SceneTensor (Söderberg et al.”
MethodNovel visual object descriptor using SURF and clustering algorithms
Grycuk, R. (2016) · Journal of Applied Mathematics and Computational Mechanics · doi:10.17512/jamcm.2016.3.04 · Cited by 5
“It was presented for the first time in [10], but currently is used in various systems e.”
BackgroundVegeShop Tool: A Tool for Vegetable Recognition Using DNN
Sakai, Y., Oda, T., Ikeda, M., et al. (2016) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-3-319-49106-6_68 · Cited by 5
Real-time panorama image synthesis by fast camera pose estimation
Kim, B. S., Lee, S. H., & Cho, N. I. (2012) · Cited by 5
Dense mapping for monocular-SLAM
Aguilar‐González, A., & Arias-Estrada, M. (2016) · International Conference on Indoor Positioning and Indoor Navigation · doi:10.1109/ipin.2016.7743671 · Cited by 4
Periocular recognition under expression variation using Higher Order Spectral features
Algashaam, F. M. A., Nguyen, K., Boles, W., & Chandran, V. (2015) · International Conference on Image and Vision Computing New Zealand (IVCNZ) · doi:10.1109/ivcnz.2015.7761569 · Cited by 7
Adaptive Local Spatiotemporal Features from RGB-D Data for One-Shot Learning Gesture Recognition
Lin, J., Ruan, X., Yu, N., & Yang, Y. (2016) · Sensors · doi:10.3390/s16122171 · Cited by 5
Underwater non-rigid 3D shape reconstruction via structure from motion for fish ethology research
Che, R., Xu, X. W., Nian, R., et al. (2016) · doi:10.1109/oceans.2016.7761289 · Cited by 5
Development of Self-Localization Method Using Sensors on Mobile Devices
Fuse, T., & MATSUMOTO, K. (2016) · Journal of the Japan society of photogrammetry and remote sensing · doi:10.4287/jsprs.54.290 · Cited by 5
Precision instrument targeting via image registration for the Mars 2020 rover
Doran, G. B., Thompson, D. R., & Estlin, T. A. (2016) · International Joint Conference on Artificial Intelligence · Cited by 7
Real-Time Object Detection and Recognition Using Internet of Things Paradigm
Jain, S. K., & Rajankar, S. O. (2017) · International Journal of Image Graphics and Signal Processing · doi:10.5815/ijigsp.2017.01.03 · Cited by 5
Transmitting What Matters: Task-Oriented Video Composition and Compression
Andaló, F., Penatti, O. A. B., & Testoni, V. (2016) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.1109/sibgrapi.2016.019 · Cited by 5
“Other possibilities include the use of distinctive local features [17], [18], [19] or even more general descriptors [20], [21].”
BackgroundBiologically Inspired Motion Encoding for Robust Global Motion Estimation
Sarıyanidi, E., Güneş, H., & Cavallaro, A. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2651394 · Cited by 5
Enhanced bio-inspired feature extraction for embedded application
Nguyen, D. D., Elouardi, A., & Bouaziz, S. (2016) · International Conference on Control, Automation, Robotics and Vision (ICARCV) · doi:10.1109/icarcv.2016.7838652 · Cited by 7
“An alternative named Speed Up Robust Feature (SURF) was proposed in [2].”
Ali, M. M., Liu, H., Stoll, N., et al. (2016) · IEEE International Symposium on Computational Intelligence and Informatics (CINTI) · doi:10.1109/cinti.2016.7846432 · Cited by 5
Adaptive video shot detection improved by fusion of dissimilarity measures
Santos, A. C. S. E., & Pedrini, H. (2016) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2016.7844688 · Cited by 5
“Examples of such transforms include SIFT [12], SURF [13], ORB [14] and BRIEF [15].”
MethodAdaptive Polar Based Filtering Method for Image Copy-Move Forgery Detection
Bi, X., Pun, C., & Yuan, X. (2016) · IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom) · doi:10.1109/trustcom.2016.0161 · Cited by 5
“The proposed method are compared with those of the state-of-the-art CMFD algorithms, the SURF feature points based forgery detection methods [10], the block-based method [17], and the SIFT based forgery detection methods in [5] .”
MethodVisual Loop-Closure Detection Method Using Average Feature Descriptors
Kim, D.-H., & Kim, J.-H. (2014) · Advances in intelligent systems and computing · doi:10.1007/978-3-319-05582-4_10 · Cited by 5
“The descriptors of the feature points are obtained by various feature point detection algorithm, such as SIFT, SURF and ORB [11][12][13].”
MethodVisual Odometry from Two Point Correspondences and Initial Automatic Camera Tilt Calibration
Wadenbäck, M., Karlsson, M., Heyden, A., et al. (2017) · International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) · doi:10.5220/0006079903400346 · Cited by 5
Hsu, C.-C., Yang, C.-K., Chien, Y.-H., et al. (2017) · Engineering Computations · doi:10.1108/ec-05-2015-0123 · Cited by 5
Egocentric hand detection via region growth
Huang, S., Wang, W., & Lü, K. (2016) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2016.7899706 · Cited by 5
“The extraction of them is faster than SIFT [19] and SURF [20] descriptors.”
MethodTemporally subsampled detection for accurate and efficient face tracking and diarization
Le, N., Heili, A., Wu, D., & Odobez, J. (2016) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2016.7899896 · Cited by 8
“The baseline is [3], and consists of frontal detector, skin filtering, and SURF-based tracker as described in the previous experiment.”
MethodInfluentialChen, Y. (2017) · PLoS ONE · doi:10.1371/journal.pone.0176969 · Cited by 5
The Bag-of-Words Method with Dictionary Analysis by Evolutionary Algorithm
Gabryel, M., & Capizzi, G. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-59060-8_5 · Cited by 5
Onmek, Y., Triboulet, J. P., Druon, S., et al. (2017) · International Conference on Control, Automation and Robotics (ICCAR) · doi:10.1109/iccar.2017.7942725 · Cited by 5
Dynamic Bayesian Networks for semantic localization in robotics
Rubio, F., Flores, M. J., Martínez-Gómez, J., & Nicholson, A. E. (2014) · Cited by 5
Ferreira, C., Penas, S. C., Silva, J. A., & Mendonça, A. M. (2019) · IEEE Portuguese Meeting on Bioengineering · doi:10.1109/enbeng.2019.8692445 · Cited by 6
Hazard Control in Industrial Environments: A Knowledge-Vision-Based Approach
de Oliveira, C. S., Sanín, C., & Szczerbicki, E. (2017) · Advances in intelligent systems and computing · doi:10.1007/978-3-319-67223-6_23 · Cited by 5
“For the feature extraction, a range of different approaches will be tested such as Histogram of Oriented Gradients [26], Speeded Up Robust Features [27], Local Binary Patterns [28] and Haar-like wavelets [29].”
MethodVide-omics: A genomics-inspired paradigm for video analysis
Kazantzidis, I., Flórez‐Revuelta, F., Dequidt, M., Hill, N. J., & Nebel, J. (2017) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2017.10.003 · Cited by 5
2D/3D Object Recognition and Categorization Approaches for Robotic Grasping
Zrira, N., Hannat, M., Bouyakhf, E. H., & Khan, H. A. (2017) · Studies in computational intelligence · doi:10.1007/978-3-319-63754-9_26 · Cited by 7
“The Speeded-Up Robust Features (SURF) [8, 9] is based on the same steps and principles of SIFT detector [36], but it utilizes a different scheme and provides better results than those obtained with SIFT extractor.”
MethodSistem Pentautan Citra Udara Menggunakan Algoritme SURF dan Metode Reduksi Data
Hamizan, Z., & Sumiharto, R. (2017) · IJEIS (Indonesian Journal of Electronics and Instrumentation Systems) · doi:10.22146/ijeis.18240 · Cited by 5
A comparison of corner and saliency detection methods for power line detection
Yetgin, Ö. E., & Gerek, Ö. N. (2017) · International Artificial Intelligence and Data Processing Symposium (IDAP) · doi:10.1109/idap.2017.8090291 · Cited by 5
“SURF method focuses its attention to saliency structures in the image [22].”
MethodA Study on Human Recognition Using Auricle and Side View Face Images
El-Naggar, S., Abaza, A. A., & Bourlai, T. (2017) · Advanced sciences and technologies for security applications · doi:10.1007/978-3-319-68533-5_4 · Cited by 6
A 3D vision based object grasping posture learning system for home service robots
Huang, Y.-L., Huang, S.-P., Chen, H., et al. (2017) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2017.8123032 · Cited by 5
“Among these methods, Scale-Invariant Feature Transform (SIFT) [1], Speeded-Up Robust Features (SURF) [2], Harris corner detector [3], and Features from Accelerated Segment Test (FAST) [4] are the most popular algorithms in recent years.”
MethodAdding Uncertainty to an Object Detection System for Mobile Robots
Hernandez, A. C., Gómez, C., Crespo, J., & Barber, R. (2017) · IEEE International Conference on Space Mission Challenges for Information Technology · doi:10.1109/smc-it.2017.9 · Cited by 7
“Some researches present the use of SURF [3] applied in indoor scenes for object recognition using mobile robots [4].”
Localization of component lead inside a THT solder joint for solder defects classification
Fonseka, C. L. S. C., & Jayasinghe, J. A. K. S. (2017) · Journal of Achievements of Materials and Manufacturing Engineering · doi:10.5604/01.3001.0010.7033 · Cited by 5
Geometric Encoding of Color Images
Nosovskii, G. V. (2018) · Moscow University Mathematics Bulletin · doi:10.3103/s0027132218010011 · Cited by 5
Endoscopic video deblurring via synthesis
Peng, L., Liu, S., Xie, D., Zhu, S., & Zeng, B. (2017) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip.2017.8305021 · Cited by 9
“However, all these methods require high quality image features [7, 8] for the accurate motion compensation.”
BackgroundMethodBarrena, N., Sánchez, J. R., Ugarte, R. J., & Alonso, A. G. (2018) · Machine Vision and Applications · doi:10.1007/s00138-018-0914-2 · Cited by 6
“Among the most used markerless tracking methods are those that use feature descriptors: SIFT [17], FREAK [1], SURF [4,5].”
MethodEvaluation of Local Features for Structure from Motion
Cao, M., Li, C., Jia, W., Li, Y., Lv, Z., Zheng, L., & Liu, X. (2018) · Multimedia Tools and Applications · doi:10.1007/s11042-018-5864-1 · Cited by 5
“Those detectors with main direction include SIFT [23] and its variants [16, 28], SURF [4], learned invariant feature transform (LIFT) [51], Oriented Fast and Rotated Brief (ORB) [32], Binary Robust Invariant Scalable Keypoints (BRISK) [18], Maximally Stable Extremal Regions (MSER) [11], and KAZE [2], etc.”
MethodFace detection based on probability of amplitude distribution of local binary patterns algorithm
Alobaidi, W. H., Aziz, I. T., Jawad, T. A., et al. (2018) · International Symposium on Digital Forensic and Security (ISDFS) · doi:10.1109/isdfs.2018.8355319 · Cited by 5
“Modern face detection algorithms have benefited from feature extraction methodologies, such as scale-invariant feature transform [3], local binary patterns (LBP) [4] and their variations, fast counterparts, such as speeded-up robust features [5], and histograms of oriented gradients (HoG) [6].”
Large-scale video compression: recent advances and challenges
Tian, T., & Wang, H. (2018) · Frontiers of Computer Science · doi:10.1007/s11704-018-7304-9 · Cited by 5
“Firstly, the SURF [73] technique is employed to extract local features of Id and Ib.”
MethodDenser Trajectories of Anchor Points for Action Recognition
Nguyen, T., Feng, J., & Nguyen, K. T. T. M. (2018) · International Conference on Ubiquitous Information Management and Communication (IMCOM) · doi:10.1145/3164541.3164611 · Cited by 6
“We deploy SURF [1] matching to rectify the video frame as mentioned in [42].”
MethodSINGLE-SHOT SEMANTIC MATCHER FOR UNSEEN OBJECT DETECTION
Сергеевич, Г. В., Vizilter, Y. V., Knyaz, V. A., & Moiseenko, A. S. (2018) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xlii-2-379-2018 · Cited by 7
“Modern algorithms are grown from handcrafted features and indexing algorithms (e.g, (Lowe, 2004), (Bay et al., 2006), (Dong et al., 2010)) to methods that uses CNN for global descriptor construction (for example (Ng et al., 2015).”
MethodWeighted two-step aggregated VLAD for image retrieval
Liu, H., Zhao, Q., Mbelwa, J., et al. (2018) · The Visual Computer · doi:10.1007/s00371-018-1573-z · Cited by 5
“, SIFT [26], RootSIFT [2] or SURF [4], many image representation methods are reported to achieve remarkable breakthroughs.”
BackgroundPlace Recognition in Semi-Dense Maps
Ye, Y., Cieslewski, T., Loquercio, A., & Scaramuzza, D. (2017) · British Machine Vision Conference (BMVC) · doi:10.5244/c.31.74 · Cited by 5
Moons, B., Bankman, D. J., & Verhelst, M. (2018) · Springer eBooks · doi:10.1007/978-3-319-99223-5_1 · Cited by 7
Interactive mobile robot in a dynamic environment
Rehman, B. U., Yagfarov, R., & Klimchik, A. S. (2018) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2018.11.331 · Cited by 5
Song, S., Kim, B., Kim, J., & Yu, S. (2018) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceanskobe.2018.8559126 · Cited by 6
“The default thresholds of SURF in MATLAB is 1,000 and more features are detected as we choose lower threshold.”
MethodInfluentialA Passive Cloud Detection System for UAV: System Functions and Validation
Funk, F., & Stuetz, P. (2019) · AIAA SciTech Forum · doi:10.2514/6.2019-2076 · Cited by 5
Road Trail Classification using Color Images for Autonomous Vehicle Navigation
Islam, K. T., Wijewickrema, S. N. R., Pervez, M., & O’Leary, S. J. (2018) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta.2018.8615834 · Cited by 5
Analysing Forests Using Dense Point Clouds
Lee, D. J., Muir, W., Beeston, S., et al. (2018) · International Conference on Image and Vision Computing New Zealand (IVCNZ) · doi:10.1109/ivcnz.2018.8634651 · Cited by 5
Bringing a virtual string quartet to life
Thielen, E., Letellier, J., Sieck, J., & Thoma, A. (2018) · African Human Computer Interaction Conference (AfriCHI) · doi:10.1145/3283458.3283477 · Cited by 5
“While it is an interesting technology for the medical [8, 9, 10] as well as the architectural sector [11, 12], it is of course also often used for enter-tainment purposes.”
BackgroundA novel SIFT architecture and ASIC implementation for real time SOC application
Qasaimeh, M., Saleh, H. H., Mohammad, B. S., et al. (2019) · Analog Integrated Circuits and Signal Processing · doi:10.1007/s10470-019-01419-9 · Cited by 5
“Such algorithms, like the SURF [3] and the fast SIFT [4], have been applied without paying much attention to the extracted features’ accuracy.”
MethodSign Language Static Gestures Recognition Tool Prototype
Imashev, A. (2017) · IEEE International Conference on Application of Information and Communication Technologies (AICT) · doi:10.1109/icaict.2017.8687032 · Cited by 5
“Unlike SIFT [14], which computes histograms over the orientation of edges in localized, small-sized areas of an image, HOG[12] computes histograms on a dense grid of uniformly spaced cells, besides SIFT as SURF[15] is a patent protected descriptor, that is why it cannot be used freely.”
Graph-Based Registration and Blending for Undersea Image Stitching
Yang, X., Liu, Z., Qiao, H., et al. (2019) · Robotica · doi:10.1017/s0263574719000699 · Cited by 5
“Besides the SIFT feature, researchers have applied many types of local features to image stitching, of which some representative features include the speeded-up robust feature (SURF), 1,31 the binary robust independent elementary features (BRIEF), 6 the shape context feature, 2 etc.”
MethodTowards Direct Localization for Visual Teach and Repeat
Gridseth, M., & Barfoot, T. D. (2019) · Canadian Conference on Computer and Robot Vision · doi:10.1109/crv.2019.00021 · Cited by 8
Adaptive segmentation-based feature extraction and S-STDM watermarking method for color image
Li, M., & Yuan, X. (2019) · Neural Computing and Applications · doi:10.1007/s00521-019-04428-x · Cited by 5
“[19] proposed the speeded up robust features (SURF) detector based on the Hessian matrix.”
MethodSupport Vector based Indoor Scene Classification Technique using Different Features
Gill, J. S., & Brar, A. S. (2019) · International Conference on Electronics, Communication and Aerospace Technology (ICECA) · doi:10.1109/iceca.2019.8822153 · Cited by 5
Ruiz, C., Falcão, J. D., & Zhang, P. (2019) · Adjunct ACM International Joint Conference on Pervasive and Ubiquitous Computing and ACM International Symposium on Wearable Computers (UbiComp/ISWC) · doi:10.1145/3341162.3345604 · Cited by 5
“Histogram of Oriented Gradients (HOG) [21], Scale Invariant Feature Transform (SIFT) [9] and Speeded-Up Robust Features (SURF) [2] are examples of approaches that classify an object by a predefined set of extracted features, but they do not perform well for large numbers of target classes and…”
BackgroundFusing Deep Local and Global Features for Remote Sensing Image Scene Classification
Yan, K., Mei, S., Ma, M., et al. (2019) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss.2019.8898963 · Cited by 5
“Traditionally, methods based on handcrafted features (e.g. ORB [1], SURF [2]) play an important role in the field of remote sensing image scene classification, and Bag-of-Words (BoW) model based methods [3] * Corresponding author.”
MethodHistorical Map Applications and Processing Technologies
Chiang, Y., Duan, W., Leyk, S., et al. (2019) · Springer briefs in geography · doi:10.1007/978-3-319-66908-3_2 · Cited by 5
Wang, Y., Zhang, C., Mu, T., et al. (2019) · Optics Communications · doi:10.1016/j.optcom.2019.125101 · Cited by 5
Features Extractors Evaluation Based V-SLAM Applications
Amraoui, M., Latif, R., Elouardi, A., et al. (2019) · World Conference on Complex Systems · doi:10.1109/icocs.2019.8930787 · Cited by 5
“SURF: this is an image feature detector and descriptor, inspired by SIFT [10].”
MethodA Spatio-Temporal Multi-Scale Binary Descriptor
Xompero, A., Lanz, O., & Cavallaro, A. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.2965277 · Cited by 5
IPDDF: an improved precision dense descriptor based flow estimation
Eng, W., Koo, V. C., & Lim, T. S. (2020) · CAAI Transactions on Intelligence Technology · doi:10.1049/trit.2019.0052 · Cited by 5
“The introduction of the keypoint detector and descriptor [15, 17, 18] has led to sparse descriptor matching, followed by adaptive sparse-to-dense interpolation for optical fl ow estimation.”
MethodEnhancing Rock Painting Tour Experience with Outdoor Augmented Reality
Zhang, Q., Zhu, X., Yu, H., et al. (2019) · IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct) · doi:10.1109/ismar-adjunct.2019.00-67 · Cited by 5
“Compared with SIFT [10] and SURF [1], ORB can run in real-time with good performance, therefore it is more suitable for outdoor rock painting image recognition task.”
MethodVisible Spectrum and Infra-Red Image Matching: A New Method
Saleem, S., & Bais, A. (2020) · Applied Sciences · doi:10.3390/app10031162 · Cited by 5
“We use OpenCV implementation of SIFT, Oriented FAST and Rotated BRIEF (ORB) [26], Speeded Up Robust Features (SURF) [31], and Binary Robust Invariant Scalable Keypoints (BRISK) [27] feature points.”
MethodInfluentialDefects Detection of Catenary Suspension Device Based on Image Processing and CNN
Li, Y., Wei, D., Wei, X., Wu, K., Liang, Y., & Shen, X. Q. (2019) · IEEE Symposium Series on Computational Intelligence (SSCI) · doi:10.1109/ssci44817.2019.9003108 · Cited by 7
Zhang, S., Yu, H. L., Wang, T., & Dong, J. (2020) · Knowledge-Based Systems · doi:10.1016/j.knosys.2020.105699 · Cited by 6
“Their experiments showed that the conventional pixel-based methods such as SIFT [16] and SURF [19] can easily fail when applied with low-visibilities.”
MethodA Novel Hybrid Keypoint Detection Algorithm for Gradual Shot Boundary Detection
Karpagavalli, S., Balamurugan, V., & Kumar, S. R. (2020) · International Conference on Emerging Trends in Information Technology and Engineering (ICETITE) · doi:10.1109/ic-etite47903.2020.343 · Cited by 5
“The value of w is set as 0.9 [11].”
BackgroundMethodInfluentialPalm-print recognition based on scale invariant features
Poonia, P., Ajmera, P. K., & Bhalerao, A. (2019) · IEEE India Conference · doi:10.1109/indicon47234.2019.9029088 · Cited by 5
Towards Real-Time, Multi-View Video Stereopsis
Ke, J., Watras, A. J., Kim, J.-J., Liu, H., Jiang, H., & Hu, Y. H. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9054391 · Cited by 6
Enriching Natural Monument with User-Generated Mobile Augmented Reality Mashup
Shin, C., Hong, S., & Yoon, H. (2020) · Journal of Multimedia Information System · doi:10.33851/jmis.2020.7.1.25 · Cited by 5
“SURF is speeded-up robust feature detection algorithm based on hessian calculation [18].”
Method
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.