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 4,001–4,100 of 14,752
Towards Visual Feature Translation
Hu, J., Ji, R., Liu, H., Zhang, S., Deng, C. D., & Tian, Q. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00312 · Cited by 15
“, color histogram [15] and shape [7]) and handcrafted local descriptors [6, 20, 30, 31, 33, 39, 47, 49], such as SIFT [29] and ORB [45].”
BackgroundEgocentric Hand Detection Via Dynamic Region Growing
Huang, S., Wang, W., He, S., & Lau, R. W. H. (2017) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3152129 · Cited by 15
“The extraction of these descriptors is much faster compared with SIFT [26] and SURF [27] descriptors.”
MethodEnhancing Stereo Image Formation and Depth Map Estimation for Mastcam Images
Kwan, C., Chou, B., & Ayhan, B. (2018) · IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · doi:10.1109/uemcon.2018.8796542 · Cited by 15
“The SURF features [21] and SIFT features [22] are extracted from the two stereo images.”
MethodLoop Closure Detection Via Maximization of Mutual Information
ZHANG, G., Yan, X., & Ye, Y. (2019) · IEEE Access · doi:10.1109/access.2019.2937967 · Cited by 15
“[16] E. Rublee, V. Rabaud, K. Konolige, and G. Bradski, ‘‘ORB: An efficient alternative to SIFT or SURF,’’ in Proc.”
MethodInfluentialApproaches on Partial Face Recognition: A Literature Review
Reshma, M., & Kannan, B. (2019) · International Conference on Trends in Electronics and Informatics (ICOEI) · doi:10.1109/icoei.2019.8862783 · Cited by 15
Huang, Y., Qiao, X., Ren, P., Liu, L., Pu, C., & Chen, J. (2019) · IEEE International Conference on Distributed Computing Systems (ICDCS) · doi:10.1109/icdcs.2019.00149 · Cited by 17
“Traditional algorithms such as SIFT [3], SURF [4] and ASIFT [5] provide fast feature extraction and matching while the generalization capacity is insufficient.”
MethodA Novel RGB-D SLAM Algorithm Based on Cloud Robotics
Liu, Y., Zhang, H., & Huang, C. (2019) · Sensors · doi:10.3390/s19235288 · Cited by 15
A new approach for crack detection and sizing in nuclear reactor cores
Devereux, M., Murray, P., & West, G. M. (2019) · Nuclear Engineering and Design · doi:10.1016/j.nucengdes.2019.110464 · Cited by 16
Alrahhal, M., & Supreethi, K. P. (2019) · International Journal of Computer Vision and Image Processing · doi:10.4018/ijcvip.2020010101 · Cited by 18
SIFT-Based Visual Tracking using Optical Flow and Belief Propagation Algorithm
Biswas, B., Ghosh, S. K., Hore, M., & Ghosh, A. (2020) · The Computer Journal · doi:10.1093/comjnl/bxz155 · Cited by 16
Registration of Multimodal Remote Sensing Images Using Transfer Optimization
Yan, X., Zhang, Y., Zhang, D., et al. (2020) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2019.2963477 · Cited by 15
A novel dense descriptor based on structure tensor voting for multi-modal image matching
Lu, J., Hu, M., Dong, J., Han, S., & Su, A. (2020) · Chinese Journal of Aeronautics · doi:10.1016/j.cja.2020.02.002 · Cited by 17
All-Pass Parametric Image Registration
Zhang, X., Gilliam, C., & Blu, T. (2020) · IEEE Transactions on Image Processing · doi:10.1109/tip.2020.2984897 · Cited by 16
Major Convolutional Neural Networks in Image Classification: A Survey
Kumar, N., Kaur, N., & Gupta, D. (2020) · Lecture notes in networks and systems · doi:10.1007/978-981-15-3020-3_23 · Cited by 17
Feng, Y., Dai, F., & Zhu, H. (2015) · Journal of Civil Structural Health Monitoring · doi:10.1007/s13349-015-0117-8 · Cited by 15
Karunanayake, N., Rajasegaran, J., Gunathillake, A., Seneviratne, S., & Jourjon, G. (2020) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2020.3007260 · Cited by 22
“…using combined style and content embeddings generated from pre-trained CNNs outperforms many baseline image retrieval meth-ods including hashing, feature-based, and structural similarity based methods such as SIFT [17], SURF [18], and SSIM [19], for the task of detecting visually similar app icons.”
MethodGeometry consistency aware confidence evaluation for feature matching
Wang, L., Chen, B., Xu, P., et al. (2020) · Image and Vision Computing · doi:10.1016/j.imavis.2020.103984 · Cited by 15
Multi-layer Architecture For Storing Visual Data Based on WCF and Microsoft SQL Server Database
Grycuk, R., Gabryel, M., Scherer, R., & Voloshynovskiy, S. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-19324-3_64 · Cited by 20
RETRACTED ARTICLE: Facial marks for enhancing facial recognition after plastic surgery
Sabharwal, T., & Gupta, R. (2020) · International Journal of Information Technology · doi:10.1007/s41870-020-00566-x · Cited by 15
“Various methodologies are proposed so far for facial acknowledgement after plastic surgery which range from Principal Component Analysis (PCA), Fisher Vector Analysis (FV), Geometric Features (GF), Local Feature Analysis (LFA), Local Binary Patterns (LBP), Speeded Up Robust Features (SURF) and Neural Network based Gab…”
RetractedToward Autonomous UAV Localization via Aerial Image Registration
Wang, X., Kealy, A., Li, W., et al. (2021) · Electronics · doi:10.3390/electronics10040435 · Cited by 15
“Registration outcomes are shown in the right column on top of their reference images.”
BackgroundA Robust Road Vanishing Point Detection Adapted to the Real-world Driving Scenes
Khac, C. N., Choi, Y., Park, J. H., et al. (2021) · Sensors · doi:10.3390/s21062133 · Cited by 15
Phobos image enhancement using unpaired multi-frame acquisitions from Indian Mars Color Camera
Misra, I., Rout, L., Arya, S., Bhateja, Y., Moorthi, S. M., & Dhar, D. (2021) · Planetary and Space Science · doi:10.1016/j.pss.2021.105215 · Cited by 15
“Different detection techniques, such as Harris (Harris and Stephens, 1988), SURF (Rublee et al., 2011), ORB (Bay et al., 2006), and SIFT (Lowe, 2004) are used to solve the matching problem.”
MethodWu, R., Xu, Z., Zhang, J., & Zhang, L. (2021) · Sensors · doi:10.3390/s21072505 · Cited by 17
Unsupervised monocular visual odometry with decoupled camera pose estimation
Lin, L., Wang, W., Luo, W., Song, L., & Zhou, W. (2021) · Digital Signal Processing · doi:10.1016/j.dsp.2021.103052 · Cited by 16
Cao, Y., Dong, Y., Wang, F., Yang, J., Cao, Y., & Li, X. S. (2021) · Optics and Lasers in Engineering · doi:10.1016/j.optlaseng.2021.106634 · Cited by 17
ReS2tAC—UAV-Borne Real-Time SGM Stereo Optimized for Embedded ARM and CUDA Devices
Ruf, B., Mohrs, J., Weinmann, M., Hinz, S., & Beyerer, J. (2021) · Sensors · doi:10.3390/s21113938 · Cited by 19
“When relying on distinctive image features such as SIFT [34], SURF [35] or ORB [36] features, a unique matching between two image points can be found.”
MethodDT-SLAM: Dynamic Thresholding Based Corner Point Extraction in SLAM System
Wu, R., Pike, M., & Lee, B. G. (2021) · IEEE Access · doi:10.1109/access.2021.3092000 · Cited by 16
“Bay et al. [14] proposed Speeded Up Robust Features (SURF), a scale and rotation-invariant method, utilizing a Hessian matrix to identify feature points.”
MethodKong, X. (2021) · Remote Sensing · doi:10.3390/rs13163152 · Cited by 18
Visual SLAM for Asteroid Relative Navigation
Dor, M., Skinner, K. A., Tsiotras, P., & Driver, T. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00235 · Cited by 15
“We denote Cocaud et al. [9], who initially lever-aged SURF [2] visual cues and range measurements, and Cocaud et al. who focused on image feature only formulation, and solved using a Rao-Blackwellized particle fil-ter [10].”
BackgroundScene image classification based on visual words concatenation of local and global features
Shrinivasa, S. R., & Prabhakar, C. J. (2021) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11354-5 · Cited by 15
“Applying restrictions (e.g., the number of divisions inside the square region) to the regular descriptor (i.e., SURF-64) f v results in several extended versions of SURF, such as SURF-36, SURF-128, and U-SURF [4, 31].”
BackgroundMethodA novel image‐orientation feature extraction method for partial discharges
Li, Z., Qian, Y., Wang, H., et al. (2021) · IET Generation Transmission & Distribution · doi:10.1049/gtd2.12356 · Cited by 15
Stereo Vision System for Vision-Based Control of Inspection-Class ROVs
Hożyń, S., & Żak, B. (2021) · Remote Sensing · doi:10.3390/rs13245075 · Cited by 15
Hierarchical Modified Fast R-CNN for Object Detection
Chaudhuri, A. (2021) · Informatica · doi:10.31449/inf.v45i7.3732 · Cited by 22
A new SURF-based algorithm for robust registration of multimodal images data
Anzid, H., Goïc, G. L., Bekkari, A., Mansouri, A., & Mammass, D. (2022) · The Visual Computer · doi:10.1007/s00371-022-02435-z · Cited by 15
A study of multi-target image-based displacement measurement approach for field testing of bridges
Colombani, I. A., & Andrawes, B. O. (2022) · Journal of Structural Integrity and Maintenance · doi:10.1080/24705314.2022.2088071 · Cited by 16
“An example of the feature identification algorithm used in this study is the popular SURF method presented by Bay et al. (2006) used for detecting distinct features in each image.”
MethodRFVTM: A Recovery and Filtering Vertex Trichotomy Matching for Remote Sensing Image Registration
Zhao, M., An, B., Wu, Y., Luong, H. V., & Kaup, A. (2016) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2016.2606899 · Cited by 15
“histograms as SIFT-like descriptors, speeded-up robust feature [27] utilizes the sums of Haar wavelet responses and estimates”
MethodSelf-Calibrating Anomaly and Change Detection for Autonomous Inspection Robots
Salimpour, S., Queralta, J. P., & Westerlund, T. (2022) · IEEE International Conference on Robotic Computing (IRC) · doi:10.1109/irc55401.2022.00042 · Cited by 18
A feature-matching method for side-scan sonar images based on nonlinear scale space
Ye, X.-F., Li, P., Zhang, J., Shi, J., & Guo, S. (2015) · Journal of Marine Science and Technology · doi:10.1007/s00773-015-0330-5 · Cited by 15
“The PCA-SIFT algorithm [8] and the speed up robust feature (SURF) algorithm [9] were proposed to overcome this problem.”
MethodCSDD Features: Center-Surround Distribution Distance for Feature Extraction and Matching
Collins, R. T., & Ge, W. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-88690-7_11 · Cited by 16
Visual odometry - A review of approaches
Zhao, B., Hu, T., & Shen, L. (2015) · IEEE International Conference on Information and Automation (ICIA) · doi:10.1109/icinfa.2015.7279718 · Cited by 15
Palmprint Based Verification System Using SURF Features
Srinivas, B. G., & Gupta, P. (2009) · Communications in computer and information science · doi:10.1007/978-3-642-03547-0_24 · Cited by 15
Telefonica Research at TRECVID 2010 Content-Based Copy Detection
Younessian, E., Anguera, X., Adamek, T., et al. (2010) · Cited by 15
A Simple Visual Navigation System with Convergence Property
Krajník, T., & Přeučil, L. (2008) · Springer eBooks · doi:10.1007/978-3-540-78317-6_29 · Cited by 15
A path planning algorithm based on fusing lane and obstacle map
Zhu, H., Fu, M., Yang, Y., Wang, X., & Wang, M. (2014) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2014.6957889 · Cited by 15
Measurement of Size and Distance of Objects Using Mobile Devices
Laotrakunchai, S., Wongkaew, A., & Patanukhom, K. (2013) · International Conference on Signal-Image Technology & Internet-Based Systems (SITIS) · doi:10.1109/sitis.2013.36 · Cited by 15
“Then, a vertical disparity of the target object between two images is determined by extracting the SURF keypoints and descriptors [7] in the target area of the first image and whole area of the second image and matching the keypoints from both image.”
MethodBounding Box Object Localization Based On Image Superpixelization
Ibrahim, M. S., Badr, A. A., Abdallah, M. R., et al. (2012) · Procedia Computer Science · doi:10.1016/j.procs.2012.09.119 · Cited by 15
A SLAM-based approach for underwater mapping using AUVs with poor inertial information
Hammond, M., & Rock, S. M. (2014) · IEEE/OES Autonomous Underwater Vehicles · doi:10.1109/auv.2014.7054419 · Cited by 18
“Some well-known algorithms that use this idea are Harris Corners [16], SIFT [17], and SURF [18].”
MethodEstimating the body portion of CT volumes by matching histograms of visual words
Feulner, J., Zhou, S. K., Seifert, S., Cavallaro, A., Hornegger, J., & Comaniciu, D. (2009) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.810240 · Cited by 18
“For all sample points inside the patient, a 72 dimensional feature vector is computed, which consists of an eight bin histogram of the Hounsfield units and a 64 dimensional oriented SURF descriptor.(7) SURF descriptors were used because good results were reported for scene classification with SIFT features.”
MethodA research of feature-based image mosaic algorithm
Yang, L., Wu, X., Zhai, J., et al. (2011) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp.2011.6100279 · Cited by 15
Stereoscopic visualization of laparoscope image using depth information from 3D model
Kumar, A., Wang, Y., Wu, C.-J., Liu, K., & Wu, H. (2014) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2013.12.013 · Cited by 17
“The SURF is faster than the other available algorithms for feature detection [28] and is very promising for the near real-time applications [29].”
BackgroundMethodInfluentialPerson Re-Identification Using Group Information
Bialkowski, A., Lucey, P., Wei, X., & Sridharan, S. (2013) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta.2013.6691512 · Cited by 17
“Gheissari et al. [2] segmented a person into regions using a triangulated graph model to compare colour and edgel information between corresponding parts of individuals, while Hamdoun et al. [3] used interest points based on a variant of SURF [4].”
MethodImproving the Accuracy of Least-Squares Probabilistic Classifiers
Yamada, M., Sugiyama, M., Wichern, G., et al. (2011) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.e94.d.1337 · Cited by 15
Image-based pose estimation for 3-D modeling in rapid, hand-held motion
Strobl, K. H., Mair, E., & Hirzinger, G. (2011) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2011.5979944 · Cited by 18
“We implemented a SURF-based relocalization stage to register back to former KLT features [27].”
MethodGPU-boosted online image matching
Chariot, A., & Keriven, R. (2008) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2008.4761174 · Cited by 17
“Some work have been proposed to improve SIFT running time: a GPU-based SIFT implementation[12] and Speeded Up Robust Features (SURF)[3].”
Audio-Based Video Genre Identification
Rouvier, M., Oger, S., Linarès, G., et al. (2015) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2014.2387411 · Cited by 15
Hierarchical Joint Max-Margin Learning of Mid and Top Level Representations for Visual Recognition
Löbel, H., Vidal, R., & Soto, Á. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.213 · Cited by 16
“As in [31], we assume that visual descriptors [18, 1] are extracted from images, either centered at interest points or by using a dense sampling scheme, and that each of these descriptors has size T .”
Lee, K. Y., Lee, J. Y., Kong, S. G., et al. (2013) · Forensic Science International · doi:10.1016/j.forsciint.2013.11.003 · Cited by 15
“Among many feature extraction algorithms [13–15] that find feature points invariant to scale and illumination, the scale-invariant feature transform (SIFT) algorithm [16] and the speeded-up robust features (SURF) algorithm [17] have been widely used.”
MethodUsing Multi-view Recognition and Meta-data Annotation to Guide a Robot's Attention
Thomas, A., Ferrari, V., Leibe, B., Tuytelaars, T., & Gool, L. V. (2009) · The International Journal of Robotics Research · doi:10.1177/0278364909340444 · Cited by 17
“This is a similar method as used in interest point detectors (Lowe, 2004; Bay et al., 2006) to determine continuous scale co-ordinates and orientations from discrete values.”
MethodSelf-citationA random center surround bottom up visual attention model useful for salient region detection
Vikram, T. N., Tscherepanow, M., & Wrede, B. (2011) · Proceedings · doi:10.1109/wacv.2011.5711499 · Cited by 15
A VISUAL NAVIGATION SYSTEM FOR UAS BASED ON GEO-REFERENCED IMAGERY
Conte, G. L., & Doherty, P. (2012) · 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-xxxviii-1-c22-101-2011 · Cited by 15
Lee, A., Lee, J.-H., & Kim, J. (2015) · ETRI Journal · doi:10.4218/etrij.16.0115.0610 · Cited by 15
Virtual 3D city model as a priori information source for vehicle localization system
Dawood, M., Cappelle, C., Najjar, M. E. B. E., Khalil, M. A., Hassan, B. E., Pomorski, D., & Peng, J. (2015) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2015.12.003 · Cited by 16
Automated in-core image generation from video to aid visual inspection of nuclear power plant cores
Murray, P., West, G. M., Marshall, S., et al. (2016) · Nuclear Engineering and Design · doi:10.1016/j.nucengdes.2015.11.037 · Cited by 15
“integrating the use of a-priori knowledge about the camera pose and position to overcome issues of the largely featureless surface areas of the graphite core which restricts the application of more traditional image stitching techniques [3-8] 2.”
MethodClassification of clothing with weighted SURF and local binary patterns
Surakarin, W., & Chongstitvatana, P. (2015) · International Computer Science and Engineering Conference · doi:10.1109/icsec.2015.7401409 · Cited by 17
“We are improving classification performance of clothing by being more selective in the subarea of the clothes combine with weighted SURF and LBP descriptor.”
MethodInfluentialRamasinghe, S., & Rodrigo, R. P. (2015) · IAPR Asian Conference on Pattern Recognition (ACPR) · doi:10.1109/acpr.2015.7486474 · Cited by 15
A 135-frames/s 1080p 87.5-mW Binary-Descriptor-Based Image Feature Extraction Accelerator
Zhu, W., Liu, L., Jiang, G., Yin, S., & Wei, S. (2015) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2015.2469116 · Cited by 15
“It achieves 1542 frames/s/W in the power efficiency, while nearly six times improvement is achieved with respect to the SURF-based accelerators proposed in [7].”
BackgroundInfluentialComplex Event Detection Using Joint Max Margin and Semantic Features
Abbasnejad, I., Sridharan, S., Denman, S., Fookes, C., & Lucey, S. (2016) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta.2016.7797023 · Cited by 18
“One popular strategy for event detection is to extract visual features such as SIFT [1] or SURF [2] from the video frames, then pool or average them over the entire video in order to represent the video as a fixed dimensional vector and finally apply a linear classifier on this fixed dimensional representation .”
MethodA Location-Aware Architecture for an IoT-Based Smart Museum
Fiore, G. D., Mainetti, L., Mighali, V., et al. (2016) · International Journal of Electronic Government Research · doi:10.4018/ijegr.2016040103 · Cited by 15
Traffic COngestion pattern classification using multi-class SVM
Nguyen, H. N., Krishnakumari, P., Vu, H. L., & van Lint, H. (2016) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2016.7795687 · Cited by 15
Automated comparison of X-ray images for cargo scanning
Visser, W., Schwaninger, A., Hardmeier, D., et al. (2016) · IEEE International Carnahan Conference on Security Technology (ICCST) · doi:10.1109/ccst.2016.7815714 · Cited by 15
“Next, a set of descriptors is computed for both the reference and the current image using SURF [21] (Speeded-Up Robust Features).”
Tan, W., Yan, B., & Lin, C. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2017.2742243 · Cited by 14
Saha, S., Xiao, D., Frost, S. M., et al. (2018) · Journal of Medical Systems · doi:10.1007/s10916-018-0911-z · Cited by 15
Dense Stereo Matching With Edge-Constrained Penalty Tuning
Chuang, T., Ting, H.-W., & Jaw, J. (2018) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2018.2805916 · Cited by 15
“For each image pair, edges are first detected by using edge drawing algorithm [16], and the pixels belonging to detected edges are matched based on the upright speeded-up robust features (U-SURF) descriptor [17].”
MethodA Deep Learning based Food Recognition System for Lifelog Images
Nguyen, B. T., Dang‐Nguyen, D., Dang, T., et al. (2018) · International Conference on Pattern Recognition Applications and Methods · doi:10.5220/0006749006570664 · Cited by 15
“The first is by using hand-crafted features (HOG Dalal and Triggs (2005), SIFT Lowe (2004), SURF Bay et al. (2006)) with two traditional machine learning models (SVM+XGBoost).”
MethodInfluentialFast and Reliable Computational Rephotography on Mobile Device
Shi, Y., Tian, F.-P., Miao, D., et al. (2018) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2018.8486559 · Cited by 15
“RANSAC-based reliable matching Although feature such as SIFT [12], SURF [13] are more robust, their computation time are unacceptably high on mobile device (>10s).”
MethodSingh, C., & Singh, J. (2019) · Information Sciences · doi:10.1016/j.ins.2019.01.058 · Cited by 17
Vision for Robust Robot Manipulation
Martínez-Martín, E., & del Pobil, A. P. (2019) · Sensors · doi:10.3390/s19071648 · Cited by 14
Wu, J., Gan, Z., Guo, W., Yang, X., & Lin, A. (2019) · Neurocomputing · doi:10.1016/j.neucom.2018.10.101 · Cited by 17
Parallel K Nearest Neighbor Matching for 3D Reconstruction
Cao, M., Li, L., Xie, W., Jia, W., Lv, Z., Zheng, L., & Liu, X. (2019) · IEEE Access · doi:10.1109/access.2019.2912647 · Cited by 19
“[53] propose to use SURF having fast speed [54]”
MethodHandgun Detection in Single-Spectrum Multiple X-ray Views Based on 3D Object Recognition
Riffo, V., Godoy, I., & Mery, D. (2019) · Journal of Nondestructive Evaluation · doi:10.1007/s10921-019-0602-9 · Cited by 16
“…obtaining highly discriminative and local invariant features related to lighting conditions and local geometric constraints (see, for example, [37] for a good review and evaluation of descriptors including the well-known SIFT [38] and SURF [39] features) or texture features (see for example [40]).”
BackgroundAutomatic Classification Using Machine Learning for Non-Conventional Vessels on Inland Waters
Włodarczyk-Sielicka, M., & Połap, D. (2019) · Sensors · doi:10.3390/s19143051 · Cited by 15
Exploring the Distribution Patterns of Flickr Photos
Ding, X. X., & Fan, H. (2019) · ISPRS International Journal of Geo-Information · doi:10.3390/ijgi8090418 · Cited by 15
Ruiz, C., Falcão, J. D., Pan, S., Noh, H. Y., & Zhang, P. (2019) · ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (BuildSys) · doi:10.1145/3360322.3360834 · Cited by 15
Robotic Grasping Using Semantic Segmentation and Primitive Geometric Model Based 3D Pose Estimation
Sun, G.-J., & Lin, H. (2020) · IEEE/SICE International Symposium on System Integration · doi:10.1109/sii46433.2020.9026297 · Cited by 14
“The most well-known descriptors such as SIFT [15] and SURF [1] find the feature points in spatial scale and extract the position, scale, and orientation.”
MethodSketchDesc: Learning Local Sketch Descriptors for Multi-View Correspondence
Deng, Y., Li, L., Zheng, Y., Lau, M., Song, Y.-Z., Chiew-lan Tai, B., & Fu, H. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3015279 · Cited by 18
Suthaharan, S., Rossi, E. A., Snyder, V. C., Chhablani, J. K., Lejoyeux, R., Sahel, J., & Dansingani, K. K. (2020) · Signal Processing · doi:10.1016/j.sigpro.2020.107733 · Cited by 13
“The Computer Vision Toolbox of Matlab also provides a commercially validated image registration, based on the intensity-based and feature-based techniques, such as speed-up robust features (SURF, [18]), KAZE – adopting the concept of wind features [19], and fast retina keypoint (FREAK, [20]).”
MethodPerformance Analysis of Hand-Crafted Features and CNN Toward Real-Time Crop Disease Identification
Tiwari, V., Agrahari, A., & Srivastava, S. (2020) · Smart innovation, systems and technologies · doi:10.1007/978-981-15-7078-0_48 · Cited by 15
Motion tracking of iris features to detect small eye movements
Chaudhary, A. K., & Pelz, J. B. (2019) · Journal of Eye Movement Research · doi:10.16910/jemr.12.6.4 · Cited by 17
Zhao, M., Zhang, G., & Ding, M. (2022) · International Journal of Remote Sensing · doi:10.1080/01431161.2021.2022240 · Cited by 17
Freegaming: Mobile, Collaborative, Adaptive and Augmented Exergaming
Görgü, L., Campbell, A. G., McCusker, K., et al. (2012) · Mobile Information Systems · doi:10.1155/2012/549173 · Cited by 15
Mobile Phone Camera-Based Video Scanning of Paper Documents
Luqman, M. M., Gomez‐Krämer, P., & Ogier, J.-M. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-05167-3_13 · Cited by 15
“They include the famous SIFT [9], SURF [10], FAST [11], ORB [12] and FREAK [13].”
MethodHybrid Visual Tracking for Augmented Books
Yang, H. S., Cho, K., Soh, J., et al. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-89222-9_18 · Cited by 15
“’s method [8], SIFT [9], and SURF [10]; however these methods consume a great deal of time to create descriptors, so they are not suitable for augmented reality systems, which require real-time performance.”
MethodFeatureless Visual Processing for SLAM in Changing Outdoor Environments
Milford, M. J., & George, A. (2013) · Springer tracts in advanced robotics · doi:10.1007/978-3-642-40686-7_38 · Cited by 16
Warp that smile on your face: Optimal and smooth deformations for face recognition
Gass, T., Pishchulin, L., Dreuw, P., & Ney, H. (2011) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg.2011.5771442 · Cited by 16
Experiments in interactive video search by addition and subtraction
Adcock, J. E., Cooper, M., & Pickens, J. (2008) · ACM International Conference on Image and Video Retrieval (CIVR) · doi:10.1145/1386352.1386412 · Cited by 17
“Finally, SURF descriptors with their vertical pixel (Yaxis) location are also computed [4] and quantized into 200 bins using online k-means [6].”
MethodThree-dimensional object recognition system for enhancing the intelligence of a KUKA robot
Xie, S. Q., Cheng, D., Wong, S., et al. (2007) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-007-1112-y · Cited by 15
Spherical Diffusion for Scale-Invariant Keypoint Detection in Wide-Angle Images
Hansen, P., Boles, W., & Corke, P. (2008) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta.2008.69 · Cited by 19
“Although numerous advances have been made relating to the problem of wide-baseline feature matching [2, 3, 18, 21, 22, 26], with the most prominent algorithm used in the literature being the Scale-Invariant Feature Transform (SIFT) [20], these methods have typically been designed for use with perspective images.”
MethodExtended Keyframe Detection with Stable Tracking for Multiple 3D Object Tracking
Park, Y.-M., Lepetit, V., & Woo, W. (2010) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2010.262 · Cited by 15
Pornographic images detection using High-Level Semantic features
Lv, L., Zhao, C., Lv, H., et al. (2011) · International Conference on Natural Computation (ICNC) · doi:10.1109/icnc.2011.6022151 · Cited by 15
“The model includes three steps: 1) Feature extraction using SURF (4 × 4) algorithm which contains robust color information and global content extraction of local features [1] [6] [10] [11] Color space that has invariant characteristic to illumination especially high brightness case is used in the proposed extended SUR…”
MethodKim, G., Oh, J., Lee, S., & Yoo, H. (2013) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2013.2253220 · Cited by 16
“First, feature descriptors from many feature extraction algorithms such as SIFT [5], Speeded Up Robust Features (SURF) [20], Binary Robust Independent Elementary Features (BRIEF) [21], and BRISK [12] contain many zeros in their descriptor values.”
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.