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 701–800 of 873
Gressin, A. (2014) · theses.fr (ABES) · doi:10.70675/10649975z075fz419dzb280zf27b8458e8e8 · Cited by 10
“La littérature abondante sur ce sujet peut être séparée entre, d’une part, la description du choix du meilleur ensemble de classifieurs [Kim et al., 2003; Malisiewicz et al., 2011] et d’autre part les méthodes de fusion des classifications ainsi obtenues [Liu et al.”
BackgroundA Novel Approach for Image Localization Using SVM Classifier and PSO Algorithm for Vehicle Tracking
Nair, E. A. C. R. (2023) · International Journal on Recent and Innovation Trends in Computing and Communication · doi:10.17762/ijritcc.v11i9.9921 · Cited by 1
Dual-perspective multi-instance embedding learning with adaptive density distribution mining
Yang, M., Tian-lin, C., Wu, W.-Z., Zeng, W.-X., Zhang, J.-Y., & Min, F. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.111063 · Cited by 1
Guo, X., Zhang, H., Liu, W., Zhang, R., & Wang, P. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3538091 · Cited by 1
“Traditional methods primarily utilized techniques such as background modeling and histograms, combined with features like color and texture [23].”
Adhinata, F. D., Wahyono, W., & Sumiharto, R. (2025) · Smart Agricultural Technology · doi:10.1016/j.atech.2025.101269 · Cited by 2
Enhance YOLOV5 Model to Detect the Ocean Ships Using Real-Time Hyperspectral Satellite Image
Dhivya, K., Venkatesan, M., & Prabhavathy, P. (2025) · Lecture notes in electrical engineering · doi:10.1007/978-981-96-1587-2_10 · Cited by 1
Yıldırım, Y., Demir, S., & Doğan, Z. (2025) · Machine learning for signal processing ... · doi:10.1109/mlsp62443.2025.11204247 · Cited by 1
“This uniform weighting makes ERM fragile in the presence of outliers, noisy data [2, 3], or class imbalance [4, 5], all of which are common in practical learning scenarios.”
Intelligent Technologies of Remote Identification of Dynamic Objects
Кондратов, О. М., Severyn, V., & Nikulina, O. (2025) · IEEE KhPI Week on Advanced Technology (KhPIWeek) · doi:10.1109/khpiweek61436.2025.11288651 · Cited by 1
“The HOG method has become a crucial foundation for many object detection methods and has found wide applications in the field of computer vision for many years [1, 6-7].”
Lightweight target detection based on improved YOLOv5n for unmanned aerial vehicle infrared imagery
Xue, W., Zhang, X., Yan, J., Chen, D., & Wang, Y. (2026) · Engineering Research Express · doi:10.1088/2631-8695/ae408c · Cited by 1
FBAO: backdoor attack against object detection via frequency noise injection
Wang, Q., Shen, H., Wang, L., Yuan, L., Ren, Y., Jia, X., Sun, S., & Meng, W. (2026) · Applied Intelligence · doi:10.1007/s10489-026-07156-7 · Cited by 1
“Clean-image [43] uses semantic information as a trigger, but this approach restricts the content of the image (e.g., many detectors [21–23].”
Building the View Graph of a Category by Exploiting Image Realism
Szabó, A. T., Vedaldi, A., & Favaro, P. (2015) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2015.109 · Cited by 1
“A less structured approach is the one of [21] that uses Exemplar SVMs to transfer viewpoint and other attributes from training images to test images.”
BackgroundOptimizing Storage Intensive Vision Applications to Device Capacity
Girdhar, R., Panda, J., & Jawahar, C. V. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16814-2_30 · Cited by 1
Evolutionary data purification for social media classification
James, S., & Collomosse, J. P. (2016) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2016.7900039 · Cited by 1
“[20] showed that training ensembles of classifiers (e.”
BackgroundBridging the experiential gap in cultural visits with computer vision
Cucchiara, R., & Bimbo, A. D. (2016) · IEEE International Forum on Research and Technology for Society and Industry (RTSI) · doi:10.1109/rtsi.2016.7740552 · Cited by 1
“The recently introduced Exemplar-SVM [5] provides a technique to recognize individual objects and add complex information to a bounding box simply transferring meta-data from the training sample to the detected object such as 3D models, so enabling augmented reality presentations.”
MethodShape-Based Pose Estimation of Robotic Surgical Instruments
Węsierski, D., & Cygert, S. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-67543-5_1 · Cited by 1
Object detection based on ensemble of exemplars
Chen, L., & Hu, F. (2015) · International Conference on Estimation, Detection and Information Fusion (ICEDIF) · doi:10.1109/icedif.2015.7280164 · Cited by 1
Eolas: Video Retrieval Application for Helping Tourists
Zhang, Z., Yang, Y., Cui, R., & Gurrin, C. G. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-04117-9_44 · Cited by 1
“Different from approach of [7] using a bag of visual word representation and a text-based indexing algorithm, we implemented a linear discriminative object classifier for each query image, as was done by [3], [6].”
MethodLearning Like a Toddler: Watching Television Series to Learn Vocabulary from Images and Audio
Yılmaz, E., Rematas, K., Tuytelaars, T., & hamme, H. V. (2014) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/2647868.2655036 · Cited by 2
“The weights w p are learned as an exemplar clas-sifier [9] but instead of a Support Vector Machine we use Linear Discriminant Analysis (LDA) [6].”
MethodEfficient Model Evaluation with Bilinear Separation Model
Xiao, F., & Hebert, M. (2015) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2015.110 · Cited by 1
Online Ensemble of Exemplar-SVMs for Visual Tracking
Chen, X., Wu, H., & Xie, X. (2012) · International Conference on Digital Home (ICDH) · doi:10.1109/icdh.2012.78 · Cited by 1
“Based on the aforementioned theories, we present in this part the framework of the proposed object tracking algorithm using an online ensemble of Exemplar-SVMs.”
BackgroundMethodInfluentialObject tracking with part-based discriminative context models
Zhu, G., Wang, J., & Lu, H. (2014) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2014.7025999 · Cited by 1
“Considering the characteristic of exemplar-SVM [23], we use it to choose discriminative parts.”
MethodDooley, D., McGinley, B., Kilmartin, L., Jones, E., Glavin, M., & Eising, C. (2014) · International Conference on Control, Automation, Robotics and Vision (ICARCV) · doi:10.1109/icarcv.2014.7064335 · Cited by 1
“From (3) and (4) it can be seen that the VOC project requires a higher degree of overlap to produce a successful detection.”
BackgroundThe any-combiner for multi-agent target classification
Parrish, N. H., & Llorens, A. J. (2013) · Cited by 1
Exemplar Network: A Generalized Mixture Model
Tsuchiya, C., Malisiewicz, T., & Torralba, A. B. (2014) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2014.113 · Cited by 1
Self-citationNearest-Neighbors Based Weighted Method for the BOVW Applied to Image Classification
Xu, M., Sun, Q., Lü, Y., & Shen, C. (2015) · Journal of Electrical Engineering and Technology · doi:10.5370/jeet.2015.10.4.1877 · Cited by 1
“As a result, instead of using AP as the metric in the comparisons, we compute the mean AP (MAP) results [26] of the models with different odd numbers from 1 to 19 in WKNN classifier to assess classification performance.”
MethodLearning to Represent Handwritten Shapes and Words for Matching and Recognition
Almazán, J. (2015) · ELCVIA Electronic Letters on Computer Vision and Image Analysis · doi:10.5565/rev/elcvia.728 · Cited by 1
Selective parts for fine-grained recognition
Li, D., Li, Y., & Wang, S. (2015) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2015.7350934 · Cited by 1
Mixture Models for Object Detection
Kuang, X., Sang, N., Chen, F., Gao, C., & Wang, R. (2015) · Communications in computer and information science · doi:10.1007/978-3-662-48558-3_32 · Cited by 1
Learning object models from few examples
Misra, I., Wang, Y.-X., & Hebert, M. (2016) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2231108 · Cited by 1
“Conventional category models are restricted in that they are highly tuned for a particular set of categories when building these models [25].”
BackgroundLearning One-Shot Exemplar SVM from the Web for Face Verification
Song, F., & Tan, X. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16811-1_27 · Cited by 1
“01 (as did in [7]) for balancing the loss penalty between the two classes.”
BackgroundMethodInfluentialMultiple instance learning for model ensemble and meta data transfer
Chen, Y., Cai, L., Zhao, Y., & Hu, F. (2016) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2016.7471998 · Cited by 1
“Unfortunately, it is hard for these frameworks to interpret object attributes till ESVM[6, 7, 8] came into being.”
BackgroundMethodTowards open-universe image parsing with broad coverage
Lazebnik, S., & Tighe, J. (2013) · International Conference on Machine Vision · Cited by 1
Multi-stage feature learning based object recognition and 3D pose estimation with kinect
Zeng, W., Liang, G., Wang, C., & Wu, X. (2016) · International Conference on Information Science and Technology · doi:10.1109/icist.2016.7483465 · Cited by 1
A flexible ensemble-SVM for computer vision tasks
Trichet, R., & O'Connor, N. E. (2016) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss.2016.7738028 · Cited by 1
“This normalization step is similar in spirit to the one performed in examplarSVMs [27] but differs by the use of the fitting function of eq.”
BackgroundMethodInfluentialSegment-based models for event detection and recounting
Kovvuri, R., Nevatia, R., & Snoek, C. G. M. (2016) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2016.7900238 · Cited by 1
“…detection scores x j = Sc ( V sj , (cid:102) M i ) are rescaled to be comparable to each other as follows: This calibration step also suppresses the responses of models that do not have high distinction in positive and negative scores by shifting the decision boundary towards the exemplars [14].”
Isolated Persian/Arabic word spotting by label embedding
Mobarakeh, M. I., & Ahmadyfard, A. (2016) · Journal of Engineering Research · doi:10.1016/s2307-1877(25)00888-0 · Cited by 1
Fast aircraft detection using cascaded discriminative model in photoelectric sensing system
Zhong, J., Leí, T., Yao, G., tang, Z., & Liu, Y. (2017) · Optical Review · doi:10.1007/s10043-017-0334-y · Cited by 1
“For one image, the average time cost is about 394 ms, which is only 1/3 of [4] and 1/17 of [2].”
BackgroundMethodInfluentialUsing layered object representation to detect partially visible airplanes in remote sensing images
Qiu, S., Wen, G., & Fan, Y. (2016) · IEEE International Conference on Progress in Informatics and Computing · doi:10.1109/pic.2016.7949486 · Cited by 1
“We compare our model with some state-of-the-art methods on the same training and testing dataset: DPM [6], ExemplarSVM [18] and Faster RCNN [19].”
MethodLend Me a Hand: Auxiliary Image Data Helps Interaction Detection
van Gemeren, C., Poppe, R. W., & Veltkamp, R. C. (2017) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg.2017.72 · Cited by 1
“We estimate parameters (αi and βi) as in [27], by measuring the ground truth overlap score of a detection with four randomly selected validation examples.”
BackgroundMethodLarge-Scale Classification of Cargo Images Using Ensemble of Exemplar-SVMs
Cheng, C., Zhang, L., & Zhang, J. (2016) · doi:10.1109/ccbd.2016.016 · Cited by 1
“For every candidate typical image patch, we train a SVM using HOG[15] feature, using itself as the only one positive sample and all other patches as the negative samples, just like the method proposed in[6], as illuminated in Figure 3.”
MethodWhat makes a place? Building bespoke place dependent object detectors for robotics
Hawke, J., Bewley, A., & Posner, I. (2017) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8206395 · Cited by 1
“Malisiewicz et al. addressed bias in intra-class variability (such as different types of buses) in general purpose image recognition with Exemplar SVMs [25], an ensemble approach which uses a nearest neighbour method to train a unique model for each positive training datum, fitted to a single positive sample and many t…”
MethodMining Mid-Level Visual Elements for Object Detection in High-Resolution Remote Sensing Images
Liu, X., Yan, H., Huo, H., & Fang, T. (2018) · IAPR Workshop on Pattern Recognition in Remote Sensing (PRRS) · doi:10.1109/prrs.2018.8486179 · Cited by 1
“V. OBJECT DETECTION USING MID-LEVEL VISUAL ELEMENTS Based on mid-level visual element representation described in Section , we train a simple one-versus-all linear SVM classifier [16] for the target detection in the particular orientation.”
MethodInfluentialSea surface object tracking for USV with spatio-temporal context fusion
Peng, Y., Chen, J., XiaoMao, L. I., Jun, L. U. O., Xie, S., Liu, C., & Pu, H. (2018) · Scientia Sinica Technologica · doi:10.1360/n092018-00218 · Cited by 1
3D model retrieval and pose estimation for indoor images by simulating scene context
Liu, M., Guo, J. J., & Guo, Y. (2019) · Graphical Models · doi:10.1016/j.gmod.2019.101032 · Cited by 1
Qin, Y., Li, B., & Lilja, D. J. (2018) · IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · doi:10.1109/uemcon.2018.8796693 · Cited by 3
“The architecture of the HPSVM proposed in this paper is built upon the Ensemble of Exemplar-SVMs (EESVMs) for object detection in [12].”
MethodRemote Sensing Image Dam Detection Based on Dual Threshold Network
Caigang, Z., Lei, Z., & Suidong, L. (2020) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc49329.2020.9164639 · Cited by 1
Development of a Search System for Similar Images
Tarasov, A. S., Tarasova, V. Y., Grinchenko, N. N., & Stepanov, M. A. (2020) · ELEKTRO · doi:10.1109/elektro49696.2020.9130343 · Cited by 1
“Therefore, it is proposed to use the Minkowski distances, namely, city-block, which is described by the following formula: . q) d(p, This formula was introduced by German Minkowski and found application in computer technology because of the high speed of calculation [5].”
MethodOnline social image ranking in diversified preferences
Zhao, X., Pei, L., Li, T., & Zhang, Z. (2020) · EURASIP Journal on Image and Video Processing · doi:10.1186/s13640-020-00540-4 · Cited by 1
A Multi-modal Audience Engagement Measurement System
Sanz, M., Masneri, S., & Zorrilla, M. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-71158-0_17 · Cited by 1
An Adaptive Video-to-Video Face Identification System Based on Self-Training
López‐López, E., Regueiro, C. V., & Pardo, X. M. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412369 · Cited by 1
“The system is inspired by the ideas of the ensembles Exemplar-SVM [29] which have been successfully used for supervised object recognition.”
MethodA REVIEW ON OBJECT DETECTION SINCE 2005
Haider, N. (2022) · International Journal of Advanced Research in Computer Science · doi:10.26483/ijarcs.v13i2.6804 · Cited by 1
A Hierarchical Cluster Tree Approach Leveraging Delaunay Triangulation
Avatavului, C., & Boiangiu, C.-A. (2023) · BRAIN BROAD RESEARCH IN ARTIFICIAL INTELLIGENCE AND NEUROSCIENCE · doi:10.18662/brain/14.3/482 · Cited by 1
“This process utilizes Non-Maximum Suppression (NMS), a technique prevalently employed in computer vision tasks to reduce the number of overlapping bounding boxes and retain the most suitable ones (Malisiewicz et al., 2011; Ren et al. 2017).”
MethodA Robust Classifier for Unbalanced-Data Learning
Zhao, L., Li, R., Ouyang, B., Wei, H., Li, Q., & Cai, H. (2023) · International Academic Exchange Conference on Science and Technology Innovation (IAECST) · doi:10.1109/iaecst60924.2023.10503228 · Cited by 1
“We first evaluate the performance of the proposed algorithms against the latest methods including Ensemble-model [8] and AUG+DenseNet201 [7], and then we compare it with four strong baseline methods: REWEI [2], SMOTE [4], HardM [1] and CostS [3].”
BackgroundMethodGradient Weighted Loss for Learning from Imbalanced Samples
Yang, Z., Zhang, X., Quan, L., & Wang, Z. (2024) · IEEE International Conference on Electronics Information and Emergency Communication · doi:10.1109/iceiec61773.2024.10561850 · Cited by 1
“To address the issue of class imbalance, algorithms like hard negative mining [12] prioritize examples that exhibit higher training losses.”
BackgroundIndexing Ensembles of Exemplar-SVMs with rejecting taxonomies
Becattini, F., Seidenari, L., & Bimbo, A. D. (2016) · International Workshop on Content-Based Multimedia Indexing (CBMI) · doi:10.1109/cbmi.2016.7500241 · Cited by 1
“Green masks in the upper row are produced using the standard ensemble [18], red masks in the lower row are generated using our hierarchical ensemble.”
MethodInfluentialDomain adaptation and generalization for visual recognition
Niu, L. (2017) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/69804 · Cited by 1
Learning a Robust Society of Tracking PartsPreprint
Burceanu, E., & Leordeanu, M. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.09602 · Cited by 1
Subspace Representations and Learning for Visual Recognition
Siyahjani, F. (2017) · The Research Repository @ WVU (West Virginia University) · doi:10.33915/etd.6652 · Cited by 1
“as exemplar SVMs(E-SVMs) [172], low rank exemplar-SVMs(LRE-SVMs) [156], discriminative sub-categorization (Sub-C) [180], unbiased metric learning (UML)[178] and methods proposed in [171] and [153], We also compare against two baseline methods of linear SVM (L-SVM) and nearest neighbour (1HNN) methods.”
Methodcvpaper.challenge in 2016: Futuristic Computer Vision through 1,600 Papers SurveyPreprint
Kataoka, H., Shirakabe, S., He, Y., Ueta, S., Suzuki, T., Abe, K., Kanezaki, A., Morita, S., Yabe, T., Kanehara, Y., et al. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1707.06436 · Cited by 1
“[1188], [1189], [1190], [1191], [1192], [1193], [1194], [1195],”
Discovering and Generating Hard Examples for Training a Red Tide Detector.Preprint
Lee, H., Kwon, H., & Kim, W. (2018) · arXiv (Cornell University) · Cited by 1
An active search strategy for efficient object class detectionPreprint
González-García, A., Vezhnevets, A. S., & Ferrari, V. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1412.3709 · Cited by 2
“Hence our approach requires the same annotation as standard object detectors [8, 10, 22, 29, 16, 40, 18, 42].”
BackgroundMethodDense Semantic Correspondence where Every Pixel is a ClassifierPreprint
Bristow, H., Valmadre, J., & Lucey, S. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1505.04143 · Cited by 1
“This is known as exemplar-based classification [12].”
Actor-Action Video Classification CSC 249/449 Spring 2020 Challenge ReportPreprint
Shi, J., Li, Z.-H., Zheng, H.-T., Xu, Y.-H., Xiao, T.-Y., Tan, W.-T., Guo, X.-N., Li, S.-Z., Yang, B., Xu, Z.-X., et al. (2020) · arXiv.org · doi:10.48550/arxiv.2008.00141 · Cited by 1
“This layer use Mish [10] as the activation function, and it is followed by a Dropout layer in order to drop redundant features and improve the model robustness.”
MethodWhy resampling outperforms reweighting for correcting sampling bias with stochastic descentsPreprint
An, J., Ying, L., & Zhu, Y. (2021) · arXiv (Cornell University) · Cited by 1
Extending Convolution to Semantic Segmentation with Graph Neural NetworksPreprint
Armstrong, S. J. N., & Grobler, H. (2023) · Research Square · doi:10.21203/rs.3.rs-3325307/v1 · Cited by 1
Self-tuned Visual Subclass Learning with Shared Samples An Incremental ApproachPreprint
Azizpour, H., & Carlsson, S. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1405.5732 · Cited by 1
“It should be noted that MEX is subtly different from other ensemble methods in that it divide the input space by training each expert on either a hard subset or non-uniform distribution of training data.”
BackgroundClassification d'images et localisation d'objets par des méthodes de type noyau de Fisher
Cinbiş, R. G. (2014) · theses.fr (ABES) · doi:10.70675/82c35162z9fc8z4a68z8a7ez0f97b0f49dc2
Detection of Aircrafts on a Collision Course using Spatio-Temporal HOG
Rozantsev, A., Christoudias, C. M., Lepetit, V., & Fua, P. (2013)
A refined object detection method based on HTM
Liu, H., Zhao, T., Wang, Y., & Tian, Y. (2014) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip.2014.7051512
“It is also the basis of a number of other state of the art algorithms including Deformable Parts Model (DPM)[2] and the Exemplar-SVM model[4].”
BackgroundGupta, N., Das, S., & Dwivedi, G. (2015) · International Conference on Perception and Machine Intelligence (PerMIn) · doi:10.1145/2708463.2709034
Per-Exemplar Fusion Learning for Video Retrieval and Recounting
Kim, I., Oh, S., Perera, A. G. A., & Lee, C. (2012) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme.2012.150
“Compared to [1, 7, 8], our method identifies relevance of features for exemplars based on discriminative analysis.”
MethodPhysical querying with multi-modal sensing
Baek, I., Stine, T., Dash, D. H., Xiao, F., Sheikh, Y. A., Movshovitz-Attias, Y., Chen, M., Hebert, M., & Takeo Kanade, A. (2014) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv.2014.6836103
“We leverage recent, complementary, advances in computer perception, such as SURF-based feature point matching and HOG-based object recognition [4, 6, 11], as well as established software tools for automatic speech recognition [10], Bayesian reasoning and systems-building [13].”
MethodExploiting domain constraints for exemplar based bus detection for traffic scheduling
Mahendran, A., Hebert, M., & Smith, S. (2014) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc.2014.6958161
“Platt calibration [26] is generally used to calibrate exemplar-SVMs.”
MethodEfficient Category Mining by Leveraging Instance Retrieval
Goel, A., Juneja, M., & Jawahar, C. V. (2013) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2013.72
“State-of-the-art methods [7, 10, 17] in object category retrieval/mining learn a SVM classifier for a set of labeled objects.”
BackgroundMethodInfluentialHough Voting with Distinctive Mid-Level Parts for Object Detection
Kuang, X., Sang, N., Chen, F., Wang, R., & Gao, C. (2014) · Communications in computer and information science · doi:10.1007/978-3-662-45646-0_31
A real-time 3D interactive system with stereo camera in the uncertain background
Yu, C. W., Chang, C., & Chen, L. (2015) · International Conference on Consumer Electronics · doi:10.1109/icce.2015.7066569
Detecting individual in crowd with moving feature's structure consistency
Yu, Y., Lei, Z., Yi, D., & Li, S. Z. (2011) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2011.6130352
“In consequence, traditional model-based techniques[15, 10, 11, 3] also fail to achieve robust and accurate result.”
BackgroundFine-Grained Image Classification Using Color Exemplar Classifiers
Zhang, C., Xiong, W., Liu, J., Zhang, Y., Liang, C., & Huang, Q. (2013) · Lecture notes in computer science · doi:10.1007/978-3-319-03731-8_31 · Cited by 1
“To generate more semantically meaningful representation, many works have been done [9-14].”
BackgroundResultExploiting Concepts In Videos For Video Event Detection
Can, E. F. (2015) · ScholarWorks@UMassAmherst (University of Massachusetts Amherst) · doi:10.7275/7362446.0
Discriminative Middle-Level Parts Mining for Object Detection
Li, D., Li, Y., & Wang, S. (2015) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2015edp7200
“Exemplar SVM [14] is an e ff ective discriminative approach to detect similar examples with a spe-cific example that is called an exemplar.”
MethodWhere was this picture painted ? - Localizing paintings by alignment to 3D models
Aubry, M., Russell, B. C., & Šivic, J. (2014) · HAL
Multi-modal video concept detection for scalable logging of pre-production broadcast content
Gray, C., Collomosse, J. P., Thorpe, J., & Turner, A. (2015) · Conference on Visual Media Production · doi:10.1145/2824840.2824841
“Recently the ensemble of exemplar-SVM (EE-SVM) framework [14] has shown greater robustness to content diversity (appearance variability) but at exponential complexity O ( e n ) in both the number of classes and training exemplars.”
BackgroundMethodInfluentialContour Detection-Based Discovery of Mid-Level Discriminative Patches for Scene Classification
Yang, J., Zhang, J., Wang, G., & Li, M. (2016) · International Journal of Advanced Robotic Systems · doi:10.5772/62266
“An exemplar SVM [21] is applied to induce a further expansion process to make more variable occurrences for each seed blocks.”
MethodCombining recognition and geometry for data-driven 3D reconstruction
Owens, A. (2013) · DSpace@MIT (Massachusetts Institute of Technology)
Hierarchical feature learning for image categorization
Zuo, Z. C. (2015) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/65659
Max-margin analysis based patch sampling for discovery of mid-level parts
Yang, L., & Xie, X. (2015) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2015.7351194
“For a given image patch e (also named as exemplar), the cost function of a SVM-like model [14, 15] is presented by”
MethodBeyond HOG: Learning Local Parts for Object Detection
Huang, C., Qin, Z., Xu, K., Wang, G., & Xu, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24075-6_28
Toward Scene Recognition by Discovering Semantic Structures and Parts
Zhou, G.-T. (2015) · Summit (Simon Fraser University)
Joint Visual Phrase Detection to Boost Scene Parsing
Tang, K., Zhao, Z., & Chen, X. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-27863-6_36
Multi-Task Regularization with Covariance Dictionary for Linear ClassifiersPreprint
Xiao, F., Luo, R., & Yu, Z. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1310.5393
“The second one is to train instance-based SVMs which means that we have only one positive sample with a large amount of negative samples (also called the “Exemplar SVM” [17]).”
BackgroundMethodFine pose estimation of known objects in cluttered scene images
Banerjee, S., Aggarwal, S., & Namboodiri, A. (2015) · Asian Conference on Pattern Recognition · doi:10.1109/acpr.2015.7486579
“We learn exemplar SVM as proposed in [13] to learn the global shape and geometric cues extracted, using the HoG and CNN features.”
MethodAlignement élastique d’images pour la reconnaissance d’objet
Duchenne, O. (2012) · theses.fr (ABES) · doi:10.70675/f2c63816zff25z49f1zb8fbze5266a51164f
Expanding the breadth and detail of object recognition
Endres, I. (2013) · Illinois Digital Environment for Access to Learning and Scholarship (University of Illinois at Urbana-Champaign) · Cited by 1
Set-Based Feature Learning for Person Re-identification via Third-Party Images
Zhao, Y., Wang, L., & Liu, Y. (2013) · Asian Conference on Pattern Recognition · doi:10.1109/acpr.2013.87
“1 gives an intuitive explanation of our third-party SVM versus Exemplar-SVM.”
MethodInfluentialIndoor Scene Classification Based on Mid-Level Features
Zhang, Q., Yang, J., & Zhang, S. (2016) · Advances in intelligent systems and computing · doi:10.1007/978-3-319-38789-5_32
Pedestrian Detection by Scene Adaptation Based on False Positive Mining
Suzuki, Y., Deguchi, D., Kawanishi, Y., Ide, I., & Murase, H. (2016) · Journal of the Japan Society for Precision Engineering · doi:10.2493/jjspe.82.1085
Latent information mining for semi-supervised sentiment classification in catering reviews
Feng, J. H., Tang, P., Feng, B., & Liu, W. (2016) · International Conference on Wireless Communications and Signal Processing (WCSP) · doi:10.1109/wcsp.2016.7752604
“Experiments [13] have shown that E-SVM is capable of extracting distinctive features for the category with less data.”
BackgroundSemantic image segmentation and cosegmentation
Zhu, H. (2015) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/62920
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.