Ensemble of exemplar-SVMs for object detection and beyond
Open accessThe DOI matches its Crossref record; 23 of the 25 references it cites were found in a registry or an index.
Abstract
This paper proposes a conceptually simple but surprisingly powerful method which combines the effectiveness of a discriminative object detector with the explicit correspondence offered by a nearest-neighbor approach. The method is based on training a separate linear SVM classifier for every exemplar in the training set. Each of these Exemplar-SVMs is thus defined by a single positive instance and millions of negatives. While each detector is quite specific to its exemplar, we empirically observe that an ensemble of such Exemplar-SVMs offers surprisingly good generalization. Our performance on the PASCAL VOC detection task is on par with the much more complex latent part-based model of Felzenszwalb et al., at only a modest computational cost increase. But the central benefit of our approach is that it creates an explicit association between each detection and a single training exemplar. Because most detections show good alignment to their associated exemplar, it is possible to transfer any available exemplar meta-data (segmentation, geometric structure, 3D model, etc.) directly onto the detections, which can then be used as part of overall scene understanding.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures


![Figure 3 Comparison. Given a bicycle training sample from PASCAL (represented with a HOG weight vector w ), we show the top 6 matches from the PASCAL test-set using three methods. Row 1: naive nearest neighbor (using raw normalized HOG). Row 2: Trained Exemplar-SVM (notice how w focuses on bike-specific edges). Row 3: Learned distance function - an Exemplar-SVM but trained in the 'distance-to-exemplar' vector space, with the exemplar being placed at the origin (loosely corresponding to [10, 14]).](https://figures.cochat.ai/paper-figures/1cc4b956c3534a219577db00fbbb8da4/ensemble-of-exemplar-svms-for-object-detection-and-beyond-figure-3.png)







Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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
“In the case of sliding-window object detectors, such as the deformable parts model (DPM) [12], this imbalance may be as extreme as 100,000 background examples to every one object.The recent trend towards object-proposal-based detectors [15,32] mitigates this issue to an extent, but the imbalance ratio may still be high (e.g., 70:1).”
From Training Region-Based Object Detectors with Online Hard Example Mining (2016) · cites this paper as [32]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
LIBSVM: A library for support vector machines
Chang, C.-C., & Lin, C.-J. (2011) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/1961189.1961199 · Cited by 40,918 · cited 1×
“Using the procedure above, we train an ensemble of Exemplar-SVM, one for each positive instance in the training set.”
MethodHistograms of Oriented Gradients for Human Detection
Dalal, N., & Triggs, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.177 · Cited by 33,584 · cited 2×
“Yet today, a number of methods can do just that with reasonable accuracy.”
BackgroundThe pascal visual object classes (VOC) challenge
Everingham, M., Van Gool, L., Williams, C. K. I., Winn, J., & Zisserman, A. (2010) · International Journal of Computer Vision · doi:10.1007/s11263-009-0275-4 · Cited by 19,599 · cited 2×
“For our experiments, we use a single source of exemplars: the PASCAL VOC 2007 dataset [8] – a popular dataset used to benchmark object detection algorithms.”
BackgroundMethodLibsvm : A library for support vector machines
Ferrari, V. (2008) · Medical Entomology and Zoology · Cited by 10,122
Object Detection with Discriminatively Trained Part-Based Models
Felzenszwalb, P. F., Girshick, R. B., McAllester, D., & Ramanan, D. (2010) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2009.167 · Cited by 9,853 · cited 4×
“For this, we present results on a set of tasks including segmentation, qualitative-geometry estimation, 3D model transfer, and related object priming.”
BackgroundMethodInfluentialProbabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods
Platt, J. C. (1999) · Cited by 6,384
Shape matching and object recognition using shape contexts
Belongie, S., Malik, J., & Puzicha, J. (2002) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.993558 · Cited by 6,306 · cited 1×
“Because most detections show good alignment to their associated exemplar, it is possible to transfer any available exemplar meta-data (segmentation, geometric structure, 3D model, etc.) directly onto the detections, which can then be used as part of overall scene understanding.”
BackgroundEstimating the Support of a High-Dimensional Distribution
Schölkopf, B., Platt, J. C., Shawe-Taylor, J., Smola, A. J., & Williamson, R. C. (2001) · Neural Computation · doi:10.1162/089976601750264965 · Cited by 6,168
SUN database: Large-scale scene recognition from abbey to zoo
Xiao, J., Hays, J., Ehinger, K. A., Oliva, A., & Torralba, A. (2010) · 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2010.5539970 · Cited by 3,283 · cited 1×
“Because most detections show good alignment to their associated exemplar, it is possible to transfer any available exemplar meta-data (segmentation, geometric structure, 3D model, etc.) directly onto the detections, which can then be used as part of overall scene understanding.”
Background80 Million Tiny Images: A Large Data Set for Nonparametric Object and Scene Recognition
Torralba, A., Fergus, R., & Freeman, W. T. (2008) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2008.128 · Cited by 1,921
Research works printed but not linked: 1
- A. Frome and J. Malik. Image retrieval and recognition using local distance functions. NIPS, 2006. 2, 3, 4
Cited by
Items 601–700 of 873
3D Object Detection Incorporating Instance Segmentation and Image Restoration
Huang, B., Huang, M., Gao, Y., Yu, Y., Jiang, X., & Zhang, J. (2019) · Wuhan University Journal of Natural Sciences · doi:10.1007/s11859-019-1407-5 · Cited by 4
Annotation-Free Word Spotting with Bag-of-Features HMMs
Rothacker, L., Wolf, F., & Fink, G. A. (2020) · International Journal of Pattern Recognition and Artificial Intelligence · doi:10.1142/s0218001421530013 · Cited by 4
Selective Learning from External Data for CT Image Segmentation
Song, Y., Yu, L., Lei, B., Choi, K., & Qin, J. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-87193-2_40 · Cited by 4
Solving The Long-Tailed Problem Via Intra- And Inter-Category Balance
Zhang, R., Lin, T., Zhang, R., & Xu, Y. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746738 · Cited by 4
Animal Tracking within a Formation of Drones
Marcos, J. T. C., & Utete, S. W. (2021) · International Conference on Information Fusion (FUSION) · doi:10.23919/fusion49465.2021.9626844 · Cited by 4
Increasing pedestrian detection performance through weighting of detection impairing factors
Hagn, K., & Grau, O. (2022) · ACM Computer Science in Cars Symposium (CSCS) · doi:10.1145/3568160.3570225 · Cited by 4
“Hard examples mining [33, 39], is another representative of sample weighting done by oversampling harder, i.”
BackgroundMerge Loss Calculation Method for Highly Imbalanced Data Multiclass Classification
Du, Z., Zhang, H., Wei, Z., Zhu, Y., Xu, J., Huang, X., & Yin, B. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3321753 · Cited by 4
“In addition, some studies focus on hard examples or assign higher weights to hard examples to obtain better model performance [18], [27], [28], [29], [32].”
libHOG: Energy-Efficient Histogram of Oriented Gradient Computation
Iandola, F., Moskewicz, M. W., & Keutzer, K. (2015) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2015.205 · Cited by 4
Video parsing via spatiotemporally analysis with images
Li, X., Mou, L., & Lu, X. (2015) · Multimedia Tools and Applications · doi:10.1007/s11042-015-2735-x · Cited by 4
“For every superpixel from the labeled superpixel pool s′i ∈ s′, training a SVM to determine its semantic label.”
BackgroundInfluentialAn Approach to Large Scale Interactive Retrieval of Cultural Heritage
Takami, M., Bell, P. R. F., & Ommer, B. (2014) · Eurographics · doi:10.2312/gch.20141307 · Cited by 4
Deep sparse representation-based mid-level visual elements discovery in fine-grained classification
Lv, L., Zhao, D., & Shao, K. (2018) · Soft Computing · doi:10.1007/s00500-018-3468-3 · Cited by 4
“Finally, we can’t obtain SVM to detect them.”
MethodInfluentialLearning detectors quickly using structured covariance matricesPreprint
Valmadre, J., Sridharan, S., & Lucey, S. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1403.7321 · Cited by 4
“The Exemplar SVM paradigm [26] involves training one linear detector per example to obtain more informative and expressive models.”
MethodGeneralized Ambiguity Decomposition for Understanding Ensemble DiversityPreprint
Audhkhasi, K., Sethy, A., Ramabhadran, B., & Narayanan, S. S. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1312.7463 · Cited by 4
“• Computer vision : Ensemble methods are often popular in computer vision tasks as well, such as tracking [14], object detection [15], and pose estimation [16].”
MethodComparison of Object Detection Algorithms on Maritime Vessels
Chua, M. Y.-K., Aha, D. W., Auslander, B., Gupta, K. M., & Morris, B. (2014) · Defense Technical Information Center (DTIC) · doi:10.21236/ada619022 · Cited by 4
“Object detection in Exemplar-SVM [2] relies on a conceptually simple but effective approach.”
MethodInfluentialContext Forest for efficient object detection with large mixture modelsPreprint
Modolo, D., Vezhnevets, A. S., & Ferrari, V. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1503.00787 · Cited by 4
“4.3) with two multi-component detectors (DPM [1] and EE-SVM [2], sec.”
BackgroundMethodInfluentialLooking out of the window: object localization by joint analysis of all windows in the image.Preprint
Vezhnevets, A. S., & Ferrari, V. (2015) · arXiv (Cornell University) · Cited by 4
Domain Adaptation and Domain Generalization with Representation Learning
Ghifary, M. (2016) · Open Access Victoria University of Wellington | Te Herenga Waka (Figshare) · doi:10.26686/wgtn.17014700.v1 · Cited by 5
Action Representation Using Classifier Decision BoundariesPreprint
Wang, J., Cherian, A., Porikli, F., & Gould, S. J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1704.01716 · Cited by 4
“In this respect, our pooling scheme is also different from Exemplar-SVMs [21, 43, 48] that learns feature filters per data sample and then use these filters for feature extraction.”
MethodStrawberry Detection Under Various Harvestation Stages
Fitter, Y. (2019) · DigitalCommons - CalPoly (California State Polytechnic University) · doi:10.15368/theses.2019.5 · Cited by 6
“#apply yellow segmentation lower = array([20,195,100]) upper = array([30,255,255]) mask = Range(hsv, lower, upper)”
BackgroundSelf‐Supervised Learning for Domain Generalization With a Multi‐Classifier Ensemble Approach
Qin, Z., Luo, Q., Nong, X., Chen, X., Zhang, H., & Wong, C. U. I. (2025) · IET Image Processing · doi:10.1049/ipr2.70098 · Cited by 3
Regularizing max-margin exemplars by reconstruction and generative models
Rubio, J. C., & Ommer, B. (2015) · 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.2015.7299049 · Cited by 3
A coarse-to-fine logo recognition method in video streams
Zhao, C., Wang, J., Xie, C., & Lu, H. (2014) · IEEE International Conference on Multimedia and Expo workshops · doi:10.1109/icmew.2014.6890576 · Cited by 3
“T. Malisiewicz et al.[20] used few or even just one positive training sample to train a detector by introducing exemplar SVM, which makes detection process more efficient, it also needs complex process to mine positive samples during training.”
BackgroundOffline learning of prototypical negatives for efficient online Exemplar SVM
Takami, M., Bell, P. R. F., & Ommer, B. (2014) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2014.6836075 · Cited by 3
“This results in a classifier for the online search in unlabeled datasets, which provides high recall in short calculation time respectively.”
BackgroundMethodShape-Based Instance Detection Under Arbitrary Viewpoint
Hsiao, E., & Hebert, M. (2013) · Advances in computer vision and pattern recognition · doi:10.1007/978-1-4471-5195-1_33 · Cited by 3
“To combine the output of view-based templates, the scores from each view are normalized [30, 38] and non-maximal suppression is applied.”
MethodEfficient Online Spatio-Temporal Filtering for Video Event Detection
Yan, X., Yuan, J., & Liang, H. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16178-5_54 · Cited by 3
“For our filtering framework, we generate a collection of prediction maps with different aspect ratios and scales with interpolated discriminative scores based on sparse outputs of Exemplar-SVMs.”
MethodInfluentialEfficient Query Specific DTW Distance for Document Retrieval with Unlimited Vocabulary
Nagendar, G., Ranjan, V., Harit, G., & Jawahar, C. V. (2018) · Journal of Imaging · doi:10.3390/jimaging4020037 · Cited by 3
“For searching complex objects in large databases, SVMs have emerged as the most popular and accurate solution in the recent past [12].”
BackgroundMethodIn-Rack Test Tube Pose Estimation Using RGB-D Data
Chen, H., Wan, W., Matsushita, M., Kotaka, T., & Harada, K. (2023) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio58561.2023.10354984 · Cited by 3
“Early object detection methods relied on handcrafted features [5][6] along with SVM classifiers [7].”
MethodBanerjee, S., Gupta, N., Das, S., Chowdhury, P. R., & Sinha, L. K. (2016) · Sadhana · doi:10.1007/s12046-016-0540-5 · Cited by 3
“Malisiewicz et al [22] use the concept of an ensemble of Exemplar-SVMs for object detection task.”
MethodExploring part-aware segmentation for fine-grained visual categorization
Pang, C., Yao, H., Sun, X., Zhao, S., & Zhang, Y. (2018) · Multimedia Tools and Applications · doi:10.1007/s11042-018-5957-x · Cited by 3
“Notice that the localization scheme is somewhat like the ensemble of exemplar svm [38], but differs in that we have more than one positive training examples in our method.”
MethodKavallieratou, E., & Likforman-Sulem, L. (2018) · MDPI eBooks · doi:10.3390/books978-3-03897-106-1 · Cited by 3
Learn to focus on objects for visual detection
Chen, Z., Li, J., & You, X. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.06.082 · Cited by 3
Group-Teaching: Learning Robust CNNs From Extremely Noisy Labels
Zheng, Y., Chen, Y., & Sarem, M. (2020) · IEEE Access · doi:10.1109/access.2020.2974779 · Cited by 3
“The idea of re-weighting each training sample has been well studied in the literature [10], [39], and [40].”
BackgroundMinimax Monte Carlo object tracking
Lim, J., Park, J.-Y., & Park, H. (2022) · The Visual Computer · doi:10.1007/s00371-022-02449-7 · Cited by 5
“Recently, various learningbased methods were proposed that can be categorized by several names such as discriminativemethods (shallow learning [4,34,47,56], deep learning [1,9,10,24,37,44,46,52,57]), reinforcement learning methods [11,23,33,55], and generative methods [15–17].”
BackgroundIMLBoost for intelligent diagnosis with imbalanced medical records
Liu, T., Chi, X., Du, Y., Yang, H., Xi, Y., & Guo, J. (2022) · Intelligent Data Analysis · doi:10.3233/ida-216050 · Cited by 3
“These studies [31,38,39] assign higher costs to hard samples compared to easy samples, but this may cause classifiers to focus on harmful samples (e.g. noisy data or mislabeled data) [34].”
Re-annotation of training samples for robust maritime object detection
Becktor, J., Boukas, E., & Nalpantidis, L. (2022) · Machine Learning with Applications · doi:10.1016/j.mlwa.2022.100411 · Cited by 4
Transfer Öğrenmeli Derin Ağlar ile İnşaat Alanında Kask Tespiti
Türkdamar, M. U., Taşyürek, M., & Öztürk, C. (2022) · Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi · doi:10.28948/ngumuh.1173944 · Cited by 3
An Appropriate Approach to Recognize Coke Size Distribution in a Blast Furnace
Wang, X., Guo, Y., & Yu, Y. (2023) · Processes · doi:10.3390/pr11010187 · Cited by 3
Nah, W. J., Ng, C. C., Lin, C.-T., Lee, Y. K., Kew, J. L., Tan, Z. Q., Chan, C. S., Zach, C., & Lai, S. (2023) · Proceedings - International Conference on Image Processing · doi:10.1109/icip49359.2023.10222779 · Cited by 3
“As such, instance-wise re-weighting methods dynamically assign loss weight to each instance during training [6,8].”
MethodEnhancement of ELDA Tracker Based on CNN Features and Adaptive Model Update
Gao, C., Shi, H., Yu, J.-G., & Sang, N. (2016) · Sensors · doi:10.3390/s16040545 · Cited by 3
“That is to say, this work follows the “HoG + ESVM” framework [20], which is quite popular in object detection or mid-level part detection.”
MethodA Novel Defect Inspection Approach Using Image Processing and Support Vector Machines in Bolts
Lv, X. F. (2015) · International Conference on Measuring Technology and Mechatronics Automation (ICMTMA) · doi:10.1109/icmtma.2015.46 · Cited by 3
“Thus, we introduce Exemplar-SVMs [8] to improve classification performance.”
MethodDiscriminative poses for early recognition in multi-camera networks
Spurlock, S., Shan, J., & Souvenir, R. (2015) · International Conference on Distributed Smart Cameras · doi:10.1145/2789116.2789117 · Cited by 3
“Methods based on Exemplar SVM [11] have been used to learn discriminative features for action recognition [6] or visual words for scene recognition [3].”
MethodAutomatic Image Synthesis from Keywords Using Scene Context
Inaba, S., Kanezaki, A., & Harada, T. (2014) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/2647868.2655009 · Cited by 3
“This approach is motivated by the fact that a classifier learned with a few positive samples and numerous negative samples tends to detect objects of the same type and orientation [10].”
BackgroundRandom clustering ferns for multimodal object recognition
Villamizar, M., Garrell, A., Sanfeliu, A., & Moreno-Noguer, F. (2016) · Neural Computing and Applications · doi:10.1007/s00521-016-2284-x · Cited by 3
“Machine learning techniques such as support vector machines [8, 10, 25, 47], boosting [1, 15, 37, 39], and more recently convolutional neural networks [9, 14, 22] are widely used to compute robust and discriminative classifiers for object detection.”
MethodLocal structured representation for generic object detection
Zhang, J., Huang, K., Tan, T. N., & Zhang, Z. (2017) · Frontiers of Computer Science · doi:10.1007/s11704-016-5530-6 · Cited by 3
Boundary Objectness Network for Object Detection and Localization
Wang, J., Tao, X., Xu, M., & Lü, J. (2018) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2018.8462434 · Cited by 3
“Over the past decade, the dominant approach to localizing the objects is the sliding window paradigm, based on HOG templates and SVM classifiers [1] [2].”
MethodUsing Adaptive Trackers for Video Face Recognition from a Single Sample Per Person
Charette-Migneault, F., Granger, É., & Mokhayeri, F. (2018) · International Conference on Image Processing Theory, Tools and Applications (IPTA) · doi:10.1109/ipta.2018.8608163 · Cited by 3
“The exemplar-SVM (e-SVM) [14] is an adaptation of SVM that applies the Different Error Costs (DEC) method [24] to counter the highly imbalanced classes.”
MethodIndoor Scene Recognition based on Weighted Voting Schemes
Hernandez, A. C., Gómez, C., Derner, E., & Barber, R. (2019) · European Conference on Mobile Robots · doi:10.1109/ecmr.2019.8870931 · Cited by 3
“The combination of classifiers has been extensively used to improve the classification results in face recognition [10], [11], emotion identification [12], texture recognition [13] and object detection [14].”
MethodAn Agent-Ensemble for Thresholded Multi-Target Classification
Parrish, N. H., Llorens, A. J., & Driskell, A. E. (2020) · Applied Sciences · doi:10.3390/app10041376 · Cited by 3
“proposed a mixture of exemplar-support vector machines (SVMs) for object classification in computer vision systems [4].”
BackgroundMethodMatching Faces and Attributes Between the Artistic and the Real Domain: the PersonArt Approach
Cornia, M., Tomei, M., Baraldi, L., & Cucchiara, R. (2022) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3490033 · Cited by 3
“[19] proposed to retrieve paintings using hand-crafted and pre-trained features either with a discriminative dimensionality reduction technique or with Exemplar SVMs [57].”
MethodDeep Reinforcement Exemplar Learning for Annotation Refinement
Li, Y., He, N., Peng, S., Ma, K., & Zheng, Y. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-87237-3_47 · Cited by 3
Semantic Image Understanding: From Pixel to Word
Fu, H. (2012) · UPT. Syiah Kuala University Library (Syiah Kuala University) · Cited by 3
Indexing quantized ensembles of exemplar-SVMs with rejecting taxonomies
Becattini, F., Seidenari, L., & Bimbo, A. D. (2017) · Multimedia Tools and Applications · doi:10.1007/s11042-017-4794-7 · Cited by 3
“In this paper we exploit taxonomies for indexing ensembles of Exemplar-SVMs (ESVM) [26], which can benefit from a hierarchical structure to reduce the computational complexity during evaluation.”
BackgroundMethodInfluentialShu, J., Xie, Q., Yi, L., Zhao, Q., Zhou, S., Xu, Z., & Meng, D. (2019) · arXiv (Cornell University) · Cited by 3
Going Deeper with Convolutional Neural Network for Intelligent Transportation
Chen, T. (2016) · Digital WPI · Cited by 4
Interactive machine vision for wildlife conservation
Kellenberger, B. (2020) · doi:10.18174/511122 · Cited by 3
Exploiting Low-rank Structure for Discriminative Sub-categorization
Xu, Z., Li, X., Yang, K., & Goldstein, T. (2015) · British Machine Vision Conference (BMVC) · doi:10.5244/c.29.149 · Cited by 2
“Exemplar-SVMs and exemplar-LDAs have been wildly adopted in visual recognition [3, 7, 21, 29, 39].”
MethodFine-tuning human pose estimations in videos
Singh, D., Balasubramanian, V. N., & Jawahar, C. V. (2016) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2016.7477680 · Cited by 2
“Hence, in this work, to overcome these issues and to strengthen the unary scores in the base model, we use the identified exemplars to train an ensemble of exemplar SVMs [10] for each part individually.”
BackgroundMethodInfluentialVisually Similar K-poselets Based Human Pose Recognition
Ni, S., Liu, W., Cheng, H., & Zhang, C. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-54526-4_31 · Cited by 2
“In this work we propose visually similar k -poselets, which solves the problem of weak and overly-generic detectors trained from visually incoherent/dissimilar categories [9].”
MethodLearning part-based spatial models for laser-vision-based room categorization
Peter, U., Leonardis, A., Skočaj, D., & Kristan, M. (2017) · The International Journal of Robotics Research · doi:10.1177/0278364917704707 · Cited by 2
Real Time Embedded Target Detection and Warning System
Fiyad, H. M. N., Metwally, H. M. B., El‐Hameed, M. A., Abozied, M. A. H., & Fiyad, H. M. N. (2020) · Journal of Physics Conference Series · doi:10.1088/1742-6596/1454/1/012007 · Cited by 2
“SL algorithms such as decision trees, boosting, support vector machines and neural network are applied to define targets [10-15].”
MethodVariational Deep Representation Learning for Cross-Modal Retrieval
Yang, C., Deng, Z., Li, T., Liu, H., & Liu, L. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-88007-1_41 · Cited by 2
Development of a Detection System for Endangered Mammals in Negros Island, Philippines Using YOLOv5n
Castañeda, J. A. J., Castro, A. L. D., Sy, M. A., AlDahoul, N., Tan, M. J. T., & Karim, H. B. A. (2023) · Lecture notes in electrical engineering · doi:10.1007/978-981-19-8406-8_35 · Cited by 2
ANALYSIS OF INFORMATION TECHNOLOGIES FOR REMOTE IDENTIFICATION OF DYNAMIC OBJECTS
Nikulina, O., Severyn, V. P., Кондратов, О. М., & Rekova, N. Y. (2023) · Bulletin of National Technical University KhPI Series System Analysis Control and Information Technologies · doi:10.20998/2079-0023.2023.01.17 · Cited by 10
Distances Based on Non-rigid Alignment for Comparison of Different Object Instances
Drayer, B., & Brox, T. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-40602-7_22 · Cited by 2
“Furthermore, we compare against the scores obtained with an exemplar SVM [12] and with a rigid alignment.”
BackgroundMethodResultInfluentialRobust object recognition in RGB-D egocentric videos based on Sparse Affine Hull Kernel
Wan, S., & Aggarwal, J. K. (2015) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2015.7301302 · Cited by 2
“Traditional object recognition has mainly focused on scenarios where the class of the object is predicted using only a single image [12, 22, 33].”
BackgroundScalable Similarity Learning Using Large Margin Neighborhood Embedding
Wang, Z., Yang, J., Lin, Z., Brandt, J., Chang, S., & Huang, T. S. (2015) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2015.68 · Cited by 2
Multi-camera head pose estimation using an ensemble of exemplars
Spurlock, S., Malmgren, P., Wu, H., & Souvenir, R. (2015) · International Conference on Distributed Smart Cameras · doi:10.1145/2789116.2789123 · Cited by 2
“For joint localization and estimation, we use an ensemble of exemplar SVMs (ESVM), which has been previously applied to object detection [8] and place recognition [4].”
MethodFast Unconstrained Vehicle Type Recognition with Dual-Layer Classification
Hu, X., Hu, B., Guo, C., & Lai, J. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-21963-9_25 · Cited by 2
“The Exemplar-SVM method [11] in objet detection domain faces the same problem since it needs to train a SVM classifier for each training sample.”
MethodObject localization with mid-level part detectors
Kuang, X., Sang, N., & Gao, C. (2016) · Neurocomputing · doi:10.1016/j.neucom.2016.02.080 · Cited by 2
Yamasaki, T., Ohshima, S., & Aizawa, K. (2015) · The Journal of The Institute of Image Information and Television Engineers · doi:10.3169/itej.69.j257 · Cited by 2
Bag of Features vs Vector of Locally Aggregated Descriptors
Younas, F., Baber, J., Mahmood, T., Farooq, J., & Bakhtyar, M. (2017) · Lecture notes in networks and systems · doi:10.1007/978-3-319-56991-8_10 · Cited by 2
Fine-grained bird recognition by using contour-based pose transfer
Zhu, L., Lv, Y., Zhang, D., Zhou, Y., Yan, G., Wang, H., & Wang, X. (2015) · Optical Engineering · doi:10.1117/1.oe.54.10.103105 · Cited by 2
“Malisiewicz and Efros11 proposed an exemplar support vector machines (SVMs) method to transfer part labels from training to test images by training a single SVM for every training image.”
BackgroundDiscovering distinctive action parts for action recognition
Chen, F., Sang, N., Gao, C., & Kuang, X. (2014) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2014.7025304 · Cited by 2
“[3] built a latent SVM model to integrate pre-selected attributes and data-driven attributes for classification.”
MethodInfluentialCoherent Occlusion Reasoning for Instance Recognition
Hsiao, E., & Hebert, M. (2013) · Asian Conference on Pattern Recognition · doi:10.1109/acpr.2013.213 · Cited by 2
“For our experiments, we fix the size of the HOG cells to be the standard 8× 8 for all the objects and learn the weights using ESVM [16]. c) Color (C): Color can be a very informative cue for object detection.”
MethodInfluentialAccelerating arrays of linear classifiers using approximate range queries
Lu, V., Endres, I., Stroila, M., & Hart, J. C. (2014) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2014.6836092 · Cited by 2
“The exemplar SVM approach [11] takes the recognition by components paradigm to an extreme by training a sepa rate object detector for each training example.”
MethodSahbi, H. (2015) · HAL · Cited by 2
Robust tracking with per‐exemplar support vector machine
Shi, R., Zhang, J., Xie, Z., Gao, J., & Zheng, X. (2015) · IET Computer Vision · doi:10.1049/iet-cvi.2014.0234 · Cited by 2
Multi-scale energy optimization for object proposal generation
Wang, C., Yang, J., Wang, K., & Lai, S. (2016) · Multimedia Tools and Applications · doi:10.1007/s11042-016-3616-7 · Cited by 2
Learning Discriminative Mid-Level Patches for Fast Scene Classification
Lin, A., Jia, X., & Chan, K. P. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-27677-9_14 · Cited by 2
Part-Aware Segmentation for Fine-Grained Categorization
Pang, C., Yao, H., Yang, Z., Sun, X., Zhao, S., & Zhang, Y. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24075-6_52 · Cited by 2
An efficient fusion approach for combining human and machine decisions
Lee, H., Kwon, H., Robinson, R. M., Nothwang, W. D., & Marathe, A. R. (2016) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2220788 · Cited by 2
Feature Selection Using Genetic Algorithms for Hand Posture Recognition
Hernandez-Belmonte, U. H., & Ayala-Ramírez, V. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-39393-3_21 · Cited by 2
Small object detection with random decision forests
Ma, J., Pan, Q., Hu, J., Zhao, C., Guo, Y., & Wang, D. (2017) · IEEE International Conference on Unmanned Systems (ICUS) · doi:10.1109/icus.2017.8278409 · Cited by 2
Wang, J., Russakovsky, O., & Ramanan, D. (2018) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2018.00199 · Cited by 2
“In the extreme case, each training example can be its own subcategory (exemplars) [31, 26], which is shown to be effective for label transfer [29, 26, 43].”
BackgroundCheng, Y., Yang, C., Chang, G.-C., & Chang, T.-W. (2018) · Multimedia Tools and Applications · doi:10.1007/s11042-018-6880-x · Cited by 2
“[21, 22] by using the samples from PASCAL VOC 2007 dataset.”
MethodLabel transfer between images and 3D shapes via local correspondence encoding
Zhu, J., Zhang, Y., Guo, J. J., Liu, H., Liu, M., Liu, Y., & Guo, Y. (2019) · Computer Aided Geometric Design · doi:10.1016/j.cagd.2019.04.009 · Cited by 2
C-FCN: Corners-based fully convolutional network for visual object detection
Jiao, L., Wang, R., & Xie, C. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-09503-3 · Cited by 2
“And it has become a cornerstone of numerous object detectors [11, 35] and been employed in various computer vision applications for many years.”
BackgroundA Trajectory-Based Method for Dynamic Scene Recognition
Peng, X., Bouzerdoum, A., & Phung, S. L. (2021) · International Journal of Pattern Recognition and Artificial Intelligence · doi:10.1142/s0218001421500294 · Cited by 2
Improving unsupervised image clustering with spatial consistency
Zhao, R., Ruan, J., Dong, B., Meng, L., & Zhang, W. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.108673 · Cited by 2
Data Discernment for Affordable Training in Medical Image Segmentation
Song, Y., Yu, L., Lei, B., Choi, K., & Qin, J. (2022) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2022.3228316 · Cited by 3
“Two main types of the techniques are data selection [10], [11], [12], [13], [14], [15], [16], [17] and data weighting [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30].”
BackgroundMethodInfluentialGuo, Z., Guan, W., & Wu, H. (2023) · Sensors · doi:10.3390/s23094542 · Cited by 2
“From 2004 onwards, many detectors were built based on this paradigm, including the HOG detector [27], DPM [28] and Exemplar SVM [29].”
MethodEfficient Media Retrieval from Non-Cooperative Queries
Shih, K. J., Di, W., Jagadeesh, V., & Piramuthu, R. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-20904-3_35 · Cited by 2
“While there is little that can be done about the shortcomings of RANSAC, it is likely that replacing the HOG filters with an exemplar detector such as with [9] or [5] could rectify the latter issue.”
BackgroundObject Detection using Dimensionality Reduction on Image Descriptors
Sharma, R. (2014) · RIT Scholar Works (Rochester Institute of Technology) · Cited by 2
Lost in the Past: Recognizing Locations Over Large Time Lags.Preprint
Fernando, B., Tommasi, T., & Tuytelaars, T. (2014) · arXiv (Cornell University) · Cited by 3
Ensemble of Part Detectors for Simultaneous Classification and LocalizationPreprint
Zhang, X., Xiong, H., Lin, W., & Tian, Q. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.10034 · Cited by 2
“Juneja et al. [20] employ an LDA accelerated version [18] of the exemplar-SVMs [26], which reduces the training cost substantially comparing with the standard SVM procedure that involves hard negative mining [15].”
InfluentialVisual-Interactive Similarity Search for Complex Objects by Example of Soccer Player AnalysisPreprint
Bernard, J., Ritter, C., Sessler, D., Zeppelzauer, M., Kohlhammer, J., & Fellner, D. W. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.03385 · Cited by 2
“Per-exemplar classification has been applied successfully for different tasks in computer vision (Malisiewicz et al., 2011).”
MethodInfluence-Balanced Loss for Imbalanced Visual ClassificationPreprint
Park, S., Lim, J., Jeon, Y., & Choi, J. Y. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.02444 · Cited by 2
“Recently, numerous studies have been conducted in which each sample is considered to design sample-wise loss functions [9, 22, 27].”
BackgroundMethodGhorbel, A. (2016) · theses.fr (ABES) · doi:10.70675/4fed0a38z8536z4f15z9b0cz4d1ac80895e6 · Cited by 2
Historical handwriting representation model dedicated to word spotting application
Wang, P. (2014) · theses.fr (ABES) · doi:10.70675/c1dca9aez8d5az4c1bz91fbzbfb022c830a8 · Cited by 2
Representing 3D models for alignment and recognition
Aubry, M. (2015) · theses.fr (ABES) · doi:10.70675/642480f0z5b8dz40b5z9d38za2c2c097f4d2 · Cited by 3
Metadata
- Authors
- Malisiewicz, T., Gupta, A., & Efros, A. A.
- Affiliations
- Carnegie Mellon University (US)
- Venue
- 2011 International Conference on Computer Vision
- Publisher
- IEEE
- Published
- 2011-11
- Pages
- 89-96
- OpenAlex
- W1989684337
- Cited by
- 865
- References
- 25
- Citations by year
- 2026: 4 · 2025: 18 · 2024: 16 · 2023: 29 · 2022: 54 · 2021: 52 · 2020: 49 · 2019: 45 · 2018: 65 · 2017: 78 · 2016: 105 · 2015: 148 · 2014: 97 · 2013: 65 · 2012: 36
- FWCI
- 48.23
- Type
- conference paper
- Language
- en
- Open access
- green
- figshare.com
- Licence
- cc-by
- Locations
- IEEE/CVF International Conference on Computer Vision (ICCV) (publishedVersion)
- Figshare (submittedVersion, cc-by)
- KiltHub Repository (submittedVersion)
- KiltHub Repository
- Funders
- National Science Foundation (0905402)
- Multidisciplinary University Research Initiative
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- object detection, linear SVM classifier, Pascal VOC, scene understanding, part-based models, generalization
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.
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.