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 8,701–8,800 of 14,752
Feature point screening for image matching
Wei, B., Liu, J., Cai, M., et al. (2021) · IEEE Conference on Industrial Electronics and Applications (ICIEA) · doi:10.1109/iciea51954.2021.9516061 · Cited by 3
Lee, H.-W., Woo, J., Roh, M.-I., Ham, S.-H., Zhao, L., Ha, S., Kim, N.-W., & Yu, C.-W. (2021) · Journal of marine science and technology · doi:10.51400/2709-6998.1583 · Cited by 4
Lai, B., Liu, W., Wang, C., et al. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-87361-5_2 · Cited by 3
Nonlinear Intensity Sonar Image Matching based on Deep Convolution Features
Zhou, X., Yu, C., Yuan, X., Wu, Y., Feng, H., & Luo, C. (2022) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceanschennai45887.2022.9775321 · Cited by 3
Tukra, S., Lidströmer, N., Ashrafian, H., & Giannarou, S. (2021) · Springer eBooks · doi:10.1007/978-3-030-58080-3_323-1 · Cited by 5
Adaptive Decision Support System for On-Line Multi-Class Learning and Object Detection
Hong, G.-J., Li, D.-L., Pare, S., Saxena, A. K., Prasad, M., & Lin, C. (2021) · Applied Sciences · doi:10.3390/app112311268 · Cited by 3
Ma, J., Xu, G., & Han, X. (2021) · IEEE Access · doi:10.1109/access.2021.3129814 · Cited by 3
“And in fact, such low-order Gaussian derivative operators have been widely applied in the applications of computer vision [28-30].”
BackgroundRadar target detection based on methods of image pattern matching
Tong, L., Heymann, F., & Noack, T. (2013) · elib (German Aerospace Center) · Cited by 3
Purposive Sample Consensus: A Paradigm for Model Fitting with Application to Visual Odometry
Wang, J., & Luo, X. (2014) · Springer tracts in advanced robotics · doi:10.1007/978-3-319-07488-7_23 · Cited by 3
“This is verified by our test results for 169 pairs of images using SURF features[17].”
MethodUP-Net: unique keyPoint description and detection net
Yang, N., Han, Y., Fang, J., Zhong, W., & Xu, A. (2021) · Machine Vision and Applications · doi:10.1007/s00138-021-01266-7 · Cited by 3
“While they show large gradients for facilitating keypoint detection [8, 9, 12, 13], many detected keypoints show similar descriptive features and cannot be well distinguished for accurate image matching.”
BackgroundSMRD: A Local Feature Descriptor for Multi-modal Image Registration
Xie, J., Jin, X., & Cao, H. (2021) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip53242.2021.9675401 · Cited by 4
Unsupervised multi-spectral image feature point matching under different lighting scenes
Huang, J., Tian, W., Wen, Y., Chen, Z., & Huang, Y. (2021) · CAA International Conference on Vehicular Control and Intelligence (CVCI) · doi:10.1109/cvci54083.2021.9661171 · Cited by 3
Applications of Deep Learning for Vehicle Detection for Smart Transportation Systems
Sharma, P., Singh, A., & Dhull, A. (2022) · Advances in intelligent systems and computing · doi:10.1007/978-981-16-6887-6_25 · Cited by 3
An Efficient BoF Representation for Object Classification
Vinoharan, V., & Ramanan, A. (2021) · ELCVIA Electronic Letters on Computer Vision and Image Analysis · doi:10.5565/rev/elcvia.1403 · Cited by 5
Moshaei-Nezhad, Y., Müller, J., Oelschlägel, M., Kirsch, M., & Tetzlaff, R. (2022) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-022-02562-x · Cited by 3
“Due to mostly low spatial resolution, low image contrast, and poor texture quality of IRT images, detecting the image features such as corners, blobs, and ridges through feature detection methods (e.g., Harris-Stephens algorithm [13], SURF [14], FAST [15], or ORB [16]) can be troublesome.”
MethodCoreView: fresh tissue biopsy assessment at the bedside using a millifluidic imaging chip
Cooper, D. J., Huang, C., Klavins, D., et al. (2022) · Lab on a Chip · doi:10.1039/d1lc01142a · Cited by 3
Selbstbestimmung, Privatheit und Datenschutz: Gestaltungsoptionen für einen europäischen Weg
Friedewald, M., Kreutzer, M., & Hansen, M. (2022) · DuD-Fachbeiträge · doi:10.1007/978-3-658-33306-5 · Cited by 6
A Lightweight Two-End Feature Fusion Network for Object 6D Pose Estimation
Zuo, L., Xie, L., Pan, H., & Wang, Z. (2022) · Machines · doi:10.3390/machines10040254 · Cited by 3
Clustering- and Transformer-Based Networks for the Style Analysis of Logo Images
Tian, N., Liu, Y., Sun, Z., & Liu, X. (2022) · Computational Intelligence and Neuroscience · doi:10.1155/2022/2090712 · Cited by 5
Reiss, M. L. L., da Rocha, R. S., Ferraz, R. S., et al. (2016) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprsarchives-xli-b5-315-2016 · Cited by 3
Mehta, A., Desai, S., & Chaturvedi, A. (2020) · International Journal of Information Systems and Informatics · doi:10.47747/ijisi.v1i1.77 · Cited by 3
Salach, A., Markiewicza, J. S., & Zawieska, D. (2016) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprsarchives-xli-b5-399-2016 · Cited by 3
Graph-Guided Fusion Penalty Based Sparse Coding for Image Classification
Yang, X., Zhang, T., Xu, C., & Xu, M. G. (2013) · Lecture notes in computer science · doi:10.1007/978-3-319-03731-8_44 · Cited by 3
“Local features, such as SIFT [15] and SURF [3], are densely extracted for a given image.”
BackgroundMonocular Localization Using Invariant Image Feature Matching to Assist Navigation
Upadhyay, V., & Balakrishnan, S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-08648-9_21 · Cited by 3
Infrared Ship Video Target Tracking Based on Cross-Connection and Spatial Transformer Network
Liu, Z., He, J., Zhang, T., Tang, R., Li, Y., & Waqas, M. M. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-06788-4_9 · Cited by 3
Humanoid robot fetching objects using monocular vision unit
Juang, L. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13602-8 · Cited by 3
“The robot vision is one of the core technologies for industry automation, object tracking detection, and artificial intelligence [3, 11, 19, 28, 30].”
BackgroundLu, Y., Zhou, J., McDorman, S. T., Zhang, C., Scott, D., Bukuts, J., Wilder, C., Smith, K. Y., & Wang, S. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-022-01669-7 · Cited by 6
Real and Pseudo Pedestrian Detection Method with CA-YOLOv5s Based on Stereo Image Fusion
Song, X., Li, G., Yang, L., Zhu, L., Hou, C., & Xiong, Z. (2022) · Entropy · doi:10.3390/e24081091 · Cited by 5
Fast Intra-Frame Video Splicing for Occlusion Removal in Diminished Reality
Lin, C., & Popescu, V. S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-16234-3_7 · Cited by 3
“R 02 is computed by finding SURF features [7] in V 01 , outside of C 1 , and in V 02 .”
MethodKongkaew, S., Ruchanurucks, M., & Takamatsu, J. (2022) · ASME Open Journal of Engineering · doi:10.1115/1.4055081 · Cited by 3
SLAM Loop Detection Algorithm Based on Improved Bag-of-Words Model
Xi, K., He, J., Hao, S., & Luo, L. (2022) · International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai55851.2022.9904167 · Cited by 3
Improved Combined Metric for Automatic Quality Assessment of Stitched Images
Okarma, K., & Kopytek, M. (2022) · Applied Sciences · doi:10.3390/app122010284 · Cited by 3
Fusion representation learning for keypoint detection and description
Sun, S., Park, U., Sun, S., & Liu, R. (2022) · The Visual Computer · doi:10.1007/s00371-022-02689-7 · Cited by 3
GH-Feat: Learning Versatile Generative Hierarchical Features From GANs
Xu, Y., Shen, Y., Zhu, J., Yang, C., & Zhou, B. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3225788 · Cited by 3
“designed features [19], [20], [21] for pattern matching and object detection.”
BackgroundFeature Consistent Point Cloud Registration in Building Information Modeling
Jiang, H., Lasang, P., Nader, G., Wu, Z., & Tanasnitikul, T. (2022) · Sensors · doi:10.3390/s22249694 · Cited by 3
An AI-Based Chicken Disease Management System
Quach, L.-D., Quoc, N. P., Thi, N. H., et al. (2022) · Lecture notes in electrical engineering · doi:10.1007/978-981-16-2183-3_68 · Cited by 3
Xiong, Q., Fang, S., Liu, X., et al. (2022) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2022.3221955 · Cited by 3
“The widely used feature matching algorithms include scale-invariant feature transform (SIFT) [2], speeded up robust feature (SURF) [3], and oriented fast and rotated brief (ORB) [4], and they are robust to geometric and illumination changes, but sensitive to NRD and texture changes.”
MethodVisual SLAM Based on Lightweight SuperPoint Network
Liu, Y., & Li, Z. (2022) · IEEE International Conference on Civil Aviation Safety and Information Technology (ICCASIT) · doi:10.1109/iccasit55263.2022.9987187 · Cited by 3
Sehli, M., Sebai, D., & Ghorbel, F. (2022) · IEEE International Workshop on Multimedia Signal Processing (MMSP) · doi:10.1109/mmsp55362.2022.9949185 · Cited by 3
“This is not to forget Speed-Up Robust Features (SURF) [20], SIFT descriptor [21], and Hough Lines method [22], etc.”
MethodLoop Closure Detection With Bidirectional Manifold Representation Consensus
Zhang, K., Li, Z., & Ma, J. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2022.3229364 · Cited by 3
“…iii) techniques used to satisfy the scalable requirement in long-term LCD. A. Image Representation 1) Handcrafted-Based Strategies: Notable early LCD approaches [23] exploit handcrafted features such as SIFT [24], SURF [25], and ORB [26], coupling with BoVW-related methods for image representation.”
MethodDrift Correction in Laboratory Nanocomputed Tomography using Joint Feature Correlation
Liu, M., Han, Y., Xi, X., Tan, S., Zhu, L., Zhang, Z., Li, L., Chen, J., & Yan, B. (2023) · Applied Optics · doi:10.1364/ao.479467 · Cited by 3
A Brief Survey of Feature Based Image Matching
Wu, X., Fu, K., Liu, Z., et al. (2022) · IEEE Conference on Industrial Electronics and Applications (ICIEA) · doi:10.1109/iciea54703.2022.10006226 · Cited by 3
“Bay et al. proposed Speeded Up Robust Features (SURF) [4].”
BackgroundMethodZhang, S., Sang, F., Li, J., Tang, T., Zhang, J., & Hou, T. (2022) · Youth Academic Annual Conference of Chinese Association of Automation (YAC) · doi:10.1109/yac57282.2022.10023860 · Cited by 3
“Later, feature-based SIFT [21] and subsequent improved algorithms [22]–[24] are proposed to make the matching more accurate, faster and more robust.”
BackgroundDeep Neural Network-based Approach for Accurate Vehicle Counting
(2023) · International journal of intelligent engineering and systems · doi:10.22266/ijies2023.0430.21 · Cited by 3
DSP-based parallel optimization for real-time video stitching
Huang, X., Tang, R., Zhou, Y., Yin, H., & Yan, C. C. (2023) · Journal of Real-Time Image Processing · doi:10.1007/s11554-023-01275-x · Cited by 5
A Visual SLAM Model Based on Lightweight SuperPoint and Depth Metric Learning
Zou, T., Duan, X., Xia, H., et al. (2023) · Lecture notes in electrical engineering · doi:10.1007/978-981-99-0479-2_134 · Cited by 3
Robust Image Matching for Camera Pose Estimation Using Oriented Fast and Rotated Brief
Bao, J., Yuan, X., & Lam, C. (2022) · International Conference on Algorithms, Computing and Artificial Intelligence (ACAI) · doi:10.1145/3579654.3579720 · Cited by 4
6D pose estimation of object based on fused region-level feature in cluttered scenes
Liu, X., Duanmu, H., An, K., Wang, W., Song, Y., Gu, Q., Yuan, B., & Wang, D. (2023) · Measurement Science and Technology · doi:10.1088/1361-6501/acc603 · Cited by 4
“…of an object from a single perspective, the feature points, such as scale-invariant feature transform (SIFT) [9], speeded-up robust features (SURF) [10], and Oriented FAST and Rotated BRIEF (ORB) [11], can be extracted from the image, and this image is matched with template images to select the…”
A Method for Building Vehicle Trajectory Data Sets Based on Drone Videos
Wang, Z., Yu, Z., Tian, W., et al. (2023) · SAE technical papers on CD-ROM/SAE technical paper series · doi:10.4271/2023-01-0714 · Cited by 3
Prozorov, D. E., & Zemtsov, A. V. (2023) · Computer Optics · doi:10.18287/2412-6179-co-1204 · Cited by 3
Accelerated and optimized covariance descriptor for pedestrian detection in self-driving cars
Abid, N., Ammari, A. C., Maashri, A. A., Abid, M., & Awadalla, M. H. A. (2023) · Design Automation for Embedded Systems · doi:10.1007/s10617-023-09273-9 · Cited by 3
“SIFT is based on the orientations of the individual gradients, whereas SURF uses the gradient information into a sub-region.”
MethodImage Hashing Robust Against Cropping and Rotation
Steinebach, M., Berwanger, T., & Liu, H. (2023) · Journal of Cyber Security and Mobility · doi:10.13052/jcsm2245-1439.1221 · Cited by 4
Chu, G., Fan, D., Dong, Y., et al. (2023) · IET Image Processing · doi:10.1049/ipr2.12821 · Cited by 3
RKHS subspace domain adaption via minimum distribution gap
Qiu, Y., Zhang, C., Xiong, C., Ma, Z., & Liao, S. (2023) · Pattern Analysis and Applications · doi:10.1007/s10044-023-01170-y · Cited by 3
“According to IGLDA [6], the SURF (Speed Up Robust Features) [33] of the dataset are first extracted; then the features are normalized and z-scored so that their means are zero and the standard deviations are set up to one.”
MethodKP2Dtiny: Quantized Neural Keypoint Detection and Description on the Edge
Rüegg, T., Giordano, M., & Magno, M. (2023) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas57966.2023.10168598 · Cited by 6
“[3] In practical applications hand crafted algorithms like SIFT [9] and its more efficient counterparts ORB [10], SURF [11] have been used for decades.”
MethodBinary Image Fingerprint: Stable Structure Identifier for 3D LiDAR Place Recognition
Zhang, G., Zhang, T., zhao, S., & Hou, L. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3297063 · Cited by 5
Adaptive Soft Encoding: A General Unsupervised Feature Aggregation Method for Place Recognition
Gong, Y., Yuan, J., Sun, F., Sun, Q., Zhu, W., & Zhang, X. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3307752 · Cited by 3
Unsupervised Deep Homography Estimation based on Transformer
Jiang, T., Fang, Q., Zhu, Q. X., et al. (2023) · International Conference on Advanced Robotics and Mechatronics (ICARM) · doi:10.1109/icarm58088.2023.10218971 · Cited by 3
“Features consist of key points and feature descriptors, and classic feature extraction methods like SIFT[14], SURF[15], ORB[16], GMS[17], and BEBLID[18] are widely used in practical scenarios.”
MethodAnalysis of Recent Trends in Image Matching Technology Based on Feature Extraction
Park, S., Chae, M. J., & Cho, S. I. (2023) · Journal of Broadcast Engineering · doi:10.5909/jbe.2023.28.4.429 · Cited by 3
A Straightforward Bifurcation Pattern-Based Fundus Image Registration Method
Ochoa-Astorga, J. E., Wang, L., Du, W., & Peng, Y. (2023) · Sensors · doi:10.3390/s23187809 · Cited by 8
“Widely used feature extractors encompass SIFT [23], SURF [33], and the Harris corner detector [34].”
MethodZhu, Z., Rowlinson, S., Chen, T., & Patching, A. (2023) · Buildings · doi:10.3390/buildings13092365 · Cited by 7
Robotic Grasping Pose Estimation based on Point Cloud accelerated by image feature correspondence
Zheng, F., Gong, Z., & Tao, B. (2023) · IEEE International Conference on Real-time Computing and Robotics (RCAR) · doi:10.1109/rcar58764.2023.10250145 · Cited by 3
Visual Tracking With Motion Distortion Removal for Nanomanipulation Inside SEM
Fu, X., Yang, Y., Sun, Z., et al. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3318726 · Cited by 3
A Hybrid Vision-Based Method of Encountered Wave Field Measurement for Navigating Surface Vehicles
Wei, H., Xiao, L., Tian, X., Zhang, B., & Lu, W. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3319028 · Cited by 3
Willot, L., Vodislav, D., Gouet-Brunet, V., Luca, L. D., & Manuel, A. (2023) · doi:10.1145/3607542.3617353 · Cited by 3
“Some of these techniques rely on learning (for instance, DELF [16]) whereas others are considered "hand-crafted": SIFT [14], SURF [2] or ORB [23], to name but a few.”
MethodA Binary Fast Image Registration Method Based on Fusion Information
Liang, H., Liu, C., Li, X., et al. (2023) · Electronics · doi:10.3390/electronics12214475 · Cited by 3
“Among them, the SURF algorithm is an improved version of the SIFT (Scale Invariant Features Transform) algorithm; it has a better registration effect and improves the detection speed of feature points [8,9].”
An Improved ORB Feature Point Extraction Algorithm based on Image Enhancement and Adaptive Threshold
Jiang, T., & Xue, Y. (2023) · IEEE Information Technology and Mechatronics Engineering Conference (ITOEC) · doi:10.1109/itoec57671.2023.10291977 · Cited by 3
MOD-IR: moving objects detection from UAV-captured video sequences based on image registration
Bouhlel, F., Mliki, H., & Hammami, M. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-16667-1 · Cited by 3
“…BRIFF (Binary Robust Independent Elementary Features) [7], RIFF (Rotation-Invariant, Fast Feature) [55], SURF (Speeded Up Robust Features) [3], ORB (Oriented FAST and Rotated BRIEF) [46], and SIFT (Scale Invariant Feature Transform) [34], have all been used as feature extraction…”
Shen, Z., Chu, H., Wang, F., Guo, Y., Liu, S., & Han, S. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-09241-1 · Cited by 4
COmparative Study of Feature Localization Methods for Endoscopy Image Matching
Urdapilleta, A., & Agudo, A. (2023) · doi:10.1109/icipc59416.2023.10328381 · Cited by 3
Improved YOLOv5 Model Based on the Pyramid Structure of Dilated Convolutions
Wu, J., & Deng, Z. (2023) · International Conference on Wavelet Analysis and Pattern Recognition · doi:10.1109/icwapr58546.2023.10337034 · Cited by 3
Background grid extraction from historical hand-drawn cadastral maps
Iftikhar, T., & Khan, N. (2023) · International Journal on Document Analysis and Recognition (IJDAR) · doi:10.1007/s10032-023-00457-4 · Cited by 3
“We compare three methods for corresponding point selection: i) manual, ii) SURF-descriptor matching [32], and iii) vertices of the background grid.”
Stereo-RSSF: stereo robust sparse scene-flow estimation
Salehi, E., Aghagolzadeh, A., & Hosseini, R. (2023) · The Visual Computer · doi:10.1007/s00371-023-03143-y · Cited by 3
“Examples of sparse SF estimation meth-ods include KLT [16], SIFT [17], SURF [18], and ORB [19] tracking.”
Zhang, Z., Song, H., Fan, J., Fu, T., Li, Q., Ai, D., Xiao, D., & Yang, J. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107890 · Cited by 3
Adhikari, P., Roy, B., Sinkar, O., Gupta, M., & Ningthoujam, C. (2023) · International Conference on Recent Advances in Science and Engineering (RAiSE) · doi:10.3390/engproc2023059121 · Cited by 3
Speeding Up SIFT for Faster Feature-Based UAV Visual Localization
Quan, N. V., Luyen, D. D., Anh, P. H., et al. (2024) · Lecture notes in networks and systems · doi:10.1007/978-3-031-50818-9_11 · Cited by 3
Guiding Local Feature Matching with Surface Curvature
Wang, S., Kannala, J., Pollefeys, M., & Baráth, D. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01648 · Cited by 15
“Traditional approaches mainly focus on improving the robustness of keypoint detection and description through, for example, extending the Harris [21] detector to handle affine transformations and multiple scales [35, 36] or using more discriminative or efficient descriptors [32, 7, 34, 1, 3].”
MethodImproved P&ID Symbol Detection Algorithm Based on YOLOv5 Network
Xiao, X. P., Li, Z., Zhao, S., Yang, L., Zhao, F., & Ge, C. (2023) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc53992.2023.10394450 · Cited by 3
Hardware Acceleration for SLAM in Mobile Systems
Fan, Z., Hao, Y.-F., Zhi, T., Guo, Q., & Du, Z. (2023) · Journal of Computer Science and Technology · doi:10.1007/s11390-021-1523-5 · Cited by 3
Dominating Set Database Selection for Visual Place Recognition
Kornilova, A., Moskalenko, I., Pushkin, T., et al. (2023) · International Conference on Advanced Robotics (ICAR) · doi:10.1109/icar58858.2023.10406721 · Cited by 3
GeoCluster: Enhancing Visual Place Recognition in Spatial Domain on Aerial Vehicle Platforms
Chen, C., He, M., Wang, J., & Meng, Z. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3363536 · Cited by 6
Calibration-Free Mobile Eye-Tracking Using Corneal Imaging
Mokatren, M., Kuflik, T., & Shimshoni, I. (2024) · Sensors · doi:10.3390/s24041237 · Cited by 3
“Other types include SURF (Sped-Up Robust Features) [23] and ORB (Oriented FAST and Rotated BRIEF) [24].”
Classification of Rice Varieties Using VGG-19 and Alexnet
Ponmalar, S., Devavaishnee, M., & Priya, M. K. I. (2023) · International Conference on Smart Generation Computing, Communication and Networking (SMART GENCON) · doi:10.1109/smartgencon60755.2023.10442412 · Cited by 3
MLV: Robust Image Matching via Multilayer Verification
Cao, M., Qi, Y., Yu, Y., Zhao, H., & Lv, Z. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3373817 · Cited by 4
“1) Feature detectors with principal directions, such as SIFT [4], speeded up robust features (SURFs) [17], KAZE [18], ORB [19], and BRISK [5]; among them, the SIFT method is a multiscale feature detector with directions and has higher repeatability, the ORB method is a binary feature detector with…”
MethodNon-parametric Representation Learning with Kernels
Esser, P., Fleissner, M., & Ghoshdastidar, D. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i11.29077 · Cited by 19
Zhao, Z., Wang, F., & You, H. (2024) · Remote Sensing · doi:10.3390/rs16071237 · Cited by 4
Özcan, H., Gençtürk, T. H., Genç, G., Yildirim, T., Durmuş, F., & Gürleyen, A. (2024) · Afyon Kocatepe University Journal of Sciences and Engineering · doi:10.35414/akufemubid.1364153 · Cited by 2
A novel deep learning based CBIR model using Convolutional Siamese Neural Networks
John, M., Mathew, T. J., & Bindu, V. R. (2024) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-219396 · Cited by 3
Efficient Feature Matching and Pose-graph Initialization for SfM
Gan, W. T., Zhan, Z. Q., & Wang, X. (2024) · 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-xlviii-1-2024-165-2024 · Cited by 5
“Firstly, in the field of image representing, some handcrafted local features such as SIFT (Lowe, 2004), SURF (Bay et al., 2006) and ORB (Rublee et al., 2011) are used to detect descriptors that is invariant for scaling, rotation, and partial invariant for viewpoint changes.”
MethodSerra, J., Lalanne, R., Dufour, J.-E., et al. (2024) · Strain · doi:10.1111/str.12479 · Cited by 3
“Some of these features might be extracted manually, but most of the information can be identified automatically using feature detection algorithms such as scale invariant feature transform (SIFT) [20] or speeded up robust features (SURF) [21].”
Nguyen, T.-H., Le, V.-H., Do, H.-S., Te, T.-H., & Phan, V.-N. (2024) · Future Internet · doi:10.3390/fi16050174 · Cited by 4
“…proposed dataset using PySLAM [15], with traditional features extracted at the feature-extraction step by feature descriptors, such as ORB [16], ORB2 [17], SIFT [4], SURF [5], BRISK [7], AKAZE [18], and KAZE [18], or features extracted from a DL model such as VGG [19], DPVO [20], and TartanVO [21].”
MethodEvaluation of 2D affine — hand-crafted detectors for feature-based TLS point cloud registration
Markiewicz, J. S. (2024) · Reports on Geodesy and Geoinformatics · doi:10.2478/rgg-2024-0008 · Cited by 6
Deep Learning Based Semantic Segmentation for BIM Model Generation from RGB-D Sensors
Rached, I., Hajji, R., Landes, T., et al. (2024) · ISPRS annals of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-annals-x-4-w5-2024-271-2024 · Cited by 3
Miao, J., Gong, S., Deng, Y., et al. (2024) · Journal of Shanghai Jiaotong University (Science) · doi:10.1007/s12204-024-2751-y · Cited by 3
Crash and Helmet Violation Detection System Using On-Device YOLOv3 Model
Surya Kumaran, S., Rokesh Kumar, P., & Hemalatha, R. (2024) · doi:10.1109/amathe61652.2024.10582119 · Cited by 3
“For detecting static objects like traffic lights, traffic signs, etc. algorithms like Histogram of Gradients [4], Haar Features, Scale Invariant Feature Transformation [5] and Speeded Up Robust Features [6] can be used.”
MethodResearch on Detection of Multiple Types of Speed Bump Defects Based on CRSCCG-YOLOv5s
Xiang, X., Chen, X., Tang, X., Luo, Q., & Ding, Y. (2024) · IEEE Access · doi:10.1109/access.2024.3446838 · Cited by 5
“…Haar-like features, Adaboost classifier, and cascade strategy), HOG features [4] (Histogram of Oriented Gradients that capture texture and shape information), and SURF [5] (Speeded-Up Robust Features, which build scale-space, detect interest points, estimate orientation, and compute descriptors).”
MethodLunar Rover Cross-View Localization Through Integration of Rover and Orbital Images
Zhao, X., Cui, L., Wei, X., Liu, C., & Yin, J. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3462487 · Cited by 4
“Traditional image registration primarily relies on feature descriptor-based [29], [30], [31], [32] and region-based [33], [34] matching methods, which exhibit stability under translation and rotation.”
LPL-VIO: monocular visual-inertial odometry with deep learning-based point and line features
Liu, C., Yang, Q., Yu, H., Fu, Q., & Akhtar, N. (2024) · Neural Computing and Applications · doi:10.1007/s00521-024-10294-z · Cited by 3
“Point feature methods Hand-crafted feature computation techniques, such as SIFT [21] and SURF [22], are widely used in robotic applications.”
FocalPose++: Focal Length and Object Pose Estimation via Render and Compare
Cífka, M., Ponimatkin, G., Labbé, Y., Russell, B. C., Aubry, M., Petrík, V., & Šivic, J. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3475638 · Cited by 8
“This task is one of the oldest problems in computer vision [1], [2], [3] and has been successfully approached by estimating the pose from 2D-3D correspondences obtained via local invariant features [1], [6], [7], [8], or by template-matching [9].”
BackgroundMethodFundamental Matrix Estimation Using Relative Depths
Ding, Y., Vávra, V., Bhayani, S., Wu, Q., Yang, J. J., & Kúkelová, Z. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73209-6_9 · Cited by 9
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.