Sampling Strategies for Bag-of-Features Image Classification
Open accessThe DOI matches its Crossref record; 26 of the 26 references it cites were found in a registry or an index.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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
“First, it is widely accepted that denser sampling of subsequences yields a better performance [24,39], so l is set to 1.”
From Comparison of Feature Learning Methods for Human Activity Recognition Using Wearable Sensors (2018) · cites this paper as [39]
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
Text categorization with Support Vector Machines: Learning with many relevant features
Joachims, T. (1998) · Lecture Notes in Computer Science · doi:10.1007/bfb0026683 · Cited by 7,999
Video Google: a text retrieval approach to object matching in videos
Sivic, & Zisserman (2003) · Proceedings Ninth IEEE International Conference on Computer Vision · doi:10.1109/iccv.2003.1238663 · Cited by 6,367 · cited 4×
“) as samplers [4, 1, 2, 5, 6, 7, 8, 9, 10, 11], but although such detectors have proven their value in matching applications, they were not designed to find the most informative patches for image classification and there is some evidence that they do not do so [12, 13].”
InfluentialThe Earth Mover's Distance as a Metric for Image Retrieval
Rubner, Y., Tomasi, C., & Guibas, L. J. (2000) · International Journal of Computer Vision · doi:10.1023/a:1026543900054 · Cited by 4,568
Visual categorization with bags of keypoints
Csurka, G. (2004) · Cited by 4,379
A comparison of affine region detectors
Mikolajczyk, K., Tuytelaars, T., Schmid, C., Zisserman, A., Matas, J., Schaffalitzky, F., Kadir, T., & Gool, L. V. (2005) · International Journal of Computer Vision · doi:10.1007/s11263-005-3848-x · Cited by 2,939
Object class recognition by unsupervised scale-invariant learning
Fergus, R., Perona, P., & Zisserman, A. (2003) · 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings. · doi:10.1109/cvpr.2003.1211479 · Cited by 2,023
Zhang, J., Marszałek, M., Lazebnik, S., & Schmid, C. (2007) · International Journal of Computer Vision · doi:10.1007/s11263-006-9794-4 · Cited by 1,923
QBIC project: querying images by content, using color, texture, and shape:
Niblack, C. W., Barber, R., Equitz, W., Flickner, M. D., Glasman, E. H., Petkovic, D., Yanker, P., Faloutsos, C., & Taubin, G. (1993) · SPIE Proceedings · doi:10.1117/12.143648 · Cited by 1,747
Representing and Recognizing the Visual Appearance of Materials Using Three-Dimensional Textons
Leung, T., & Malik, J. (2001) · International Journal of Computer Vision · doi:10.1023/a:1011126920638 · Cited by 1,498 · cited 4×
“Patches are selected using keypoints [4, 1, 2, 5, 6, 7, 8, 9, 10, 11] or densely [3, 13, 15].”
Influential
Research works printed but not linked: 1
- Niblack, W., Barber, R., Equitz, W., Flickner, M., Glasman, D., Petkovic, D., Yanker, P.: The qbic project: Querying image by content using color, texture, and shape. SPIE 1908, 173–187 (1993)
Cited by
Items 301–400 of 925
Multi‐pyramid image spatial structure based on coarse‐to‐fine pyramid and scale space
Xu, J., Wang, N., & Wang, Y. (2018) · CAAI Transactions on Intelligence Technology · doi:10.1049/trit.2018.1017 · Cited by 22
Binder, A., Samek, W., Kloft, M., & Kawanabe, M. (2011) · Fraunhofer-Publica (Fraunhofer-Gesellschaft) · Cited by 27
Large-scale multimedia data mining using MapReduce framework
Wang, H., Shen, Y. C., Wang, L., Zhufeng, K., Wang, W., & Cheng, C. (2012) · IEEE International Conference on Cloud Computing Technology and Science · doi:10.1109/cloudcom.2012.6427595 · Cited by 22
“The keypoint detector and descriptor are the dense sampling detector [13] and SIFT descriptor [15], respectively.”
MethodTowards a solid solution of real-time fire and flame detection
Jiang, B., Lu, Y., Li, X., & Lin, L. (2014) · Multimedia Tools and Applications · doi:10.1007/s11042-014-2106-z · Cited by 22
“Recent researches like [12] and [18] indicate that dense sampling gives better results than keypoint sampling in local feature based object classification.”
BackgroundLearning Optimal Compact Codebook for Efficient Object Categorization
Li, T., Mei, T., & Kweon, I. S. (2008) · Proceedings · doi:10.1109/wacv.2008.4544027 · Cited by 22
“Recent research shows that the dense sampling strategy can yield better classification performance than interest points detector [7, 13].”
BackgroundTowards Pose-Invariant 2D Face Classification for Surveillance
Sanderson, C., Shang, T., & Lovell, B. C. (2007) · Springer eBooks · doi:10.1007/978-3-540-75690-3_21 · Cited by 22
“Inspired by text classification techniques from the fields of natural language processing and information retrieval, alternative forms of the “bag of features” approach are used for image categorisation in [10–12].”
MethodImage classification using SLIC superpixel and FAAGKFCM image segmentation
Singh, N. K., Singh, N. J., & Kumar, W. K. (2019) · IET Image Processing · doi:10.1049/iet-ipr.2019.0255 · Cited by 24
“…to sample the patches [1, 5, 14, 17–26], the sampling of patches is done and they are used for clustering to give BoF which in turn uses an SVM classifier for image classification or by using densely with regular patches, in [27–30] they use a regular grid for finding the texture of the image.”
MethodCreating Efficient Visual Codebook Ensembles for Object Categorization
Luo, H., Wei, H., & Lai, L. L. (2010) · IEEE Transactions on Systems Man and Cybernetics - Part A Systems and Humans · doi:10.1109/tsmca.2010.2064300 · Cited by 22
“In the majority of the studies pertaining to the bag-of-words model, the codebook size was in the range of 100 to 1000 [24], [27], [41].”
BackgroundShape recognition for ships: World War I naval camouflage under the magnifying glass
Bekers, W., Meyer, R. D., & Strobbe, T. (2016) · WIT transactions on the built environment · doi:10.2495/dshf160151 · Cited by 21
“The third step in this chain is the use of SVM (Support Vector Machine), a set of machine learning algorithms that attempt to classify data based on similarity or difference [17, 18].”
MethodAction recognition in still images by learning spatial interest regions from videos
Eweiwi, A., Cheema, M. S., & Bauckhage, C. (2014) · Pattern Recognition Letters · doi:10.1016/j.patrec.2014.07.017 · Cited by 21
“[12] empirically demonstrate that random sampling provides equal or better activity classifiers than sophisticated multi-scale interest point detectors; yet, their work also illustrates that the most important aspect of sampling is the number of sample points extracted.”
BackgroundExploiting Feature Correlations by Brownian Statistics for People Detection and Recognition
Bąk, S., Biagio, M. S., Kumar, R., Murino, V., & Brémond, F. (2016) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2016.2531658 · Cited by 22
“has been proven to be very useful for image matching and object recognition under real-world conditions [15]–[18].”
BackgroundContextual Bag-of-Words for Robust Visual Tracking
Zeng, F., Ji, Y., & Levine, M. D. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2778561 · Cited by 23
Image Classification Based on Mid-Level Feature Fusion
Aslam, M. A., Salik, M. N., Chughtai, F., Ali, N., Dar, S. H., & Khalil, T. (2019) · International Conference on Emerging Technologies · doi:10.1109/icet48972.2019.8994721 · Cited by 25
ScGSLC: An unsupervised graph similarity learning framework for single-cell RNA-seq data clustering
Li, J., Jiang, W., Han, H., Liu, J., Liu, B., & Wang, Y. (2020) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2020.107415 · Cited by 24
Patch-Based Discriminative Learning for Remote Sensing Scene Classification
Muhammad, U., Hoque, M. Z., Wang, W., & Oussalah, M. (2022) · Remote Sensing · doi:10.3390/rs14235913 · Cited by 24
Multi-label learning with label relevance in advertising video
Hou, S., Zhou, S., Chen, L., Feng, Y., & Karim, A. (2015) · Neurocomputing · doi:10.1016/j.neucom.2015.07.022 · Cited by 24
A Tale of Two Object Recognition Methods for Mobile Robots
Arnau Ramisa, A., Vasudevan, S., Scaramuzza, D., de Mántaras, R. L., & Siegwart, R. (2008) · Springer eBooks · doi:10.1007/978-3-540-79547-6_34 · Cited by 21
“In order to make the local descriptors robust to changes in point of view and scale, local feature detectors are often used to select the image patches that will be used [5], although some authors point that using bigger numbers of randomly selected patches gives better results than a limited number of regions defined…”
BackgroundMelanoma and Other Skin Lesion Detection Using Smart Handheld Devices
Zouridakis, G., Wadhawan, T., Situ, N., Hu, R., Yuan, X., Lancaster, K., & Queen, C. M. (2014) · Methods in molecular biology · doi:10.1007/978-1-4939-2172-0_30 · Cited by 23
“To address the problem of imbalanced classes in a small sample size classification, several technique have been developed based on resampling, namely, undersampling the majority class [100, 105], oversampling the minority class [100–107], or both [108]; one-class learning methods built to recognize samples from a give…”
MethodMultimodal Web Aesthetics Assessment Based on Structural SVM and Multitask Fusion Learning
Wu, O., Zuo, H., Hu, W., & Li, B. (2016) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2016.2538722 · Cited by 22
“In the bag-of-words technique [34], local feature extraction is carried out using the following steps.”
MethodContent-Based Image Retrieval by Indexing Random Subwindows with Randomized Trees
Marée, R., Geurts, P., & Wehenkel, L. A. (2007) · Springer eBooks · doi:10.1007/978-3-540-76390-1_60 · Cited by 21
Barata, C., Marques, J. S., & Rozeira, J. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-38628-2_85 · Cited by 21
“A simple way of detecting the informative key points is assuming that each keypoint is a node of a regular grid placed on the lesion [9].”
BackgroundInfluentialSoft-assigned bag of features for object tracking
Ren, T., Qiu, Z., Liu, Y., Yu, T., & Bei, J. (2014) · Multimedia Systems · doi:10.1007/s00530-014-0384-y · Cited by 20
Automatic Facial Expression Recognition using Bags of Motion Words
Xu, L., & Mordohai, P. (2010) · British Machine Vision Conference (BMVC) · doi:10.5244/c.24.13 · Cited by 22
“The placement of descriptors at dense grids has been shown to be more effective for recognition than placing them at detected feature locations [18, 28].”
BackgroundScalable scene understanding using saliency-guided object localization
Bharath, R., Nicholas, L. Z. J., & Cheng, X. (2013) · International Conference on Control and Automation · doi:10.1109/icca.2013.6565074 · Cited by 20
“References [25], [26] also state that all the local features of the training images need not be used for dictionary learning and computational time can be limited by selecting a subset of features.”
BackgroundBalance between object and background: Object-enhanced features for scene image classification
Ji, Z., Wang, J., Su, Y., Song, Z., & Xing, S. (2013) · Neurocomputing · doi:10.1016/j.neucom.2012.02.054 · Cited by 22
Dynamic Foreground/Background Extraction from Images and Videos using Random Patches
Lü, L., & Hager, G. D. (2007) · The MIT Press eBooks · doi:10.7551/mitpress/7503.003.0121 · Cited by 20
Relaxed matching kernels for robust image comparison
Vedaldi, A., & Soatto, S. (2008) · 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.2008.4587619 · Cited by 20
“To this end, several operators (feature detectors) are available, ranging from the selection of random patches [16] to the extraction of scale or affine covariant blobs and corners [15].”
Can, T., Karalı, A. O., & Aytaç, T. (2011) · Applied Optics · doi:10.1364/ao.50.006302 · Cited by 20
“We used the k-means algorithm [25–28] to cluster the target and background SIFT descriptors by using different k values for target and background.”
MethodLarge-scale Dictionary Learning For Local Coordinate Coding
Xie, B., Song, M., & Tao, D. (2010) · British Machine Vision Conference (BMVC) · doi:10.5244/c.24.36 · Cited by 19
“Vector Quantization (VQ), or using k-means to learn data cluster centroids, is a simple and popular method in the bag-of-features framework [9].”
MethodUsing A Bag of Words for Automatic Medical Image Annotation With A Latent Semantic
Bouslimi, R., Messaoudi, A., & Akaichi, J. (2013) · International Journal of Artificial Intelligence & Applications · doi:10.5121/ijaia.2013.4305 · Cited by 19
A survey on image retrieval performance of different bag of visual words indexing techniques
Mukherjee, J., Mukhopadhyay, J., & Mitra, P. (2014) · IEEE Students’ Technology Symposium · doi:10.1109/techsym.2014.6807922 · Cited by 19
“In content based image retrieval, the BoVW based technique is found to outperform several other methods that are in the literature and it is widely used in many systems [22], [20], [14].”
MethodGuan, T., Fan, Y., Duan, L., & Yu, J. (2014) · PLoS ONE · doi:10.1371/journal.pone.0098806 · Cited by 19
“In that case, traditional image describing (such as BOF [12]) and indexing (such as inverted index) techniques commonly used in the field of CBIR (Content-based image retrieval) can be directly taken to fulfill the location recognition task.”
MethodRapid Online Analysis of Local Feature Detectors and Their Complementarity
Ehsan, S., Clark, A. F., & McDonald-Maier, K. D. (2013) · Sensors · doi:10.3390/s130810876 · Cited by 17
“However, it has been emphasized that repeatability is not the only characteristic that guarantees performance in a particular vision application [1,11], as attributes such as efficiency and the density of detected features are also important.”
BackgroundCrowdsourced test report prioritization considering bug severity
Tong, Y., & Zhang, X. (2021) · Information and Software Technology · doi:10.1016/j.infsof.2021.106668 · Cited by 20
Bubble space and place representation in topological maps
Erkent, Ö., & Bozma, H. I. (2013) · The International Journal of Robotics Research · doi:10.1177/0278364913481393 · Cited by 19
“, 2008) used together with HIST (Ulrich and Nourbakhsh, 2000) or classic BoW (Nowak et al., 2006; Wang and Lin, 2011).”
MethodResultInfluentialSearch pruning in video surveillance systems: Efficiency-reliability tradeoff
Dantcheva, A., Singh, A. K., Elia, P., & Dugelay, J. (2011) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw.2011.6130409 · Cited by 19
Imaging Informatics: Toward Capturing and Processing Semantic Information in Radiology Images
Napel, S. A., & Rubin, D. L. (2010) · Yearbook of Medical Informatics · doi:10.1055/s-0038-1638686 · Cited by 20
“One approach to capturing image semantics is analyzing and processing “visual words” in images (captured as image patches or codebooks [73-75]); these techniques have shown to perform well in CBIR applications [76].”
MethodCMIB: Unsupervised Image Object Categorization in Multiple Visual Contexts
Yan, X., Ye, Y., Qiu, X., Manic, M., & Yu, H. L. (2019) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2019.2939278 · Cited by 19
“In particular, Harries corner detection [35] is adopted to detect the local interest points, while a dense sampling method [36] is utilized to capture the dense interest points for the global image features.”
MethodLow-Power Dynamic Object Detection and Classification With Freely Moving Event Cameras
Ramesh, B., Ussa, A., Vedova, L. D., Yang, H., & Orchard, G. (2020) · Frontiers in Neuroscience · doi:10.3389/fnins.2020.00135 · Cited by 26
“Besides feature selection, larger dictionaries or codebooks tend to provide higher classification accuracy (Nowak et al., 2006), however, the high-dimensionality of the object representation when combined with spatial pyramid matching (14 times the codebook size for a 1 × 1, 2 × 2, and 3 × 3 SPM…”
BackgroundMultiple Order Gradient Feature for Macro-Invertebrate Identification Using Support Vector Machines
Tirronen, V., Caponio, A., Haanpää, T., & Meissner, K. K. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-04921-7_50 · Cited by 19
“Other popular methods include such examples as SIFT [16] which relies on gradient based local features extracted around key-points, Bag-of-Features (summarized in [ 17 ]) which creates ’code-books’ of textons found in images, and [18] where a system randomly sampling sub-images from the original is used.”
MethodHierarchical and discriminative bag of features for face profile and ear based gender classification
Zhang, G., & Wang, Y. (2011) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb.2011.6117590 · Cited by 19
“Most aspects of bag of features technique have been deeply investigated in the literature, and various algorithms have been proposed [3, 13, 19, 21].”
BackgroundLossless compression of binary image descriptors for visual sensor networks
Ascenso, J., & Pereira, F. M. B. (2013) · IEEE International Conference on Digital Signal Processing (DSP) · doi:10.1109/icdsp.2013.6622692 · Cited by 18
A survey on aggregating methods for action recognition with dense trajectories
Xu, H., Tian, Q., Wang, Z., & Wu, J. (2015) · Multimedia Tools and Applications · doi:10.1007/s11042-015-2536-2 · Cited by 19
“Because dense sampling has shown to improve results over sparse interest points for image classification [12, 36].”
BackgroundOrdered trajectories for human action recognition with large number of classes
Murthy, O. V. R., & Goecke, R. (2015) · Image and Vision Computing · doi:10.1016/j.imavis.2015.06.009 · Cited by 18
“[23] showed by extensive experiments that uniform random sampling can provide performances comparable to dense sampling of interest points.”
BackgroundSegmentation Based Interest Points and Evaluation of Unsupervised Image Segmentation Methods
Koniusz, P., & Mikolajczyk, K. (2009) · British Machine Vision Conference (BMVC) · doi:10.5244/c.23.24 · Cited by 21
“Experiments performed in [16] explain poor performance of MSER/HE.”
BackgroundMethodAction recognition by saliency-based dense sampling
Xu, Z., Hu, R., Chen, J., Chen, C., Chen, H., Li, H., & Sun, Q. (2016) · Neurocomputing · doi:10.1016/j.neucom.2016.09.106 · Cited by 19
“Since recognition task always bene fi t from dense features but sparse [60], we should replace feature points with those points sampled by original iDT after excluding minimum warped fl ow, when the sampled ratio becomes relative small.”
Deep Learning on Binary Patterns for Face Recognition
Vinay, A., Gupta, A., Bharadwaj, A., Srinivasan, A., Murthy, K. N. B., & Natarajan, S. (2018) · Procedia Computer Science · doi:10.1016/j.procs.2018.05.164 · Cited by 19
Supporting Ancient Coin Classification by Image-Based Reverse Side Symbol Recognition
Anwar, H., Zambanini, S., & Kampel, M. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-40246-3_3 · Cited by 18
Histopathological Image Classification Using Stain Component Features on a pLSA Model
Díaz, G. M., & Romero, E. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-16687-7_12 · Cited by 18
“The proposed local region detection and description methods were compared with traditional grid based and sift point detection sampling strategies [10].”
MethodDaily activity recognition combining gaze motion and visual features
Shiga, Y., Toyama, T., Utsumi, Y., Kise, K., & Dengel, A. R. (2014) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/2638728.2641691 · Cited by 18
Enhanced representation and multi-task learning for image annotation
Binder, A., Samek, W., Müller, K., & Kawanabe, M. (2012) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2012.09.006 · Cited by 20
“However too sparse sampling of key points may hurt the recognition and ranking performance substantially as shown in [32] and also supported by the good results of per pixel dense sampling in [33].”
ResultOn-the-fly hand detection training with application in egocentric action recognition
Kumar, J., Li, Q., Kyal, S., Bernal, E. A., & Bala, R. (2015) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2015.7301344 · Cited by 16
“Researchers have explored adaptive sampling for the analysis of third-person images and video in order to reduce computation and/or improve accuracy for tasks such as object and action recognition [45, 46, 47, 48, 49].”
BackgroundDynamic Amelioration of Resolution Mismatches for Local Feature Based Identity Inference
Wong, Y., Sanderson, C., Mau, S., & Lovell, B. C. (2010) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2010.299 · Cited by 17
“The MRH approach is motivated by the ‘visual words’ technique originally used in image categorisation [10].”
An extended-HCT semantic description for visual place recognition
Wang, M.-L., & Lin, H. (2011) · The International Journal of Robotics Research · doi:10.1177/0278364911406760 · Cited by 17
“Angeli et al. (2009) represented each topological location using the bag of visual words paradigm (Nowak et al. 2006; Lazebnik et al. 2006).”
MethodTOP-SIFT: the selected SIFT descriptor based on dictionary learning
Liu, Y., Deng, Y., Chen, X., Li, Z., & Jianping Fan, B. (2018) · The Visual Computer · doi:10.1007/s00371-018-1502-1 · Cited by 17
Biomedical Imaging Informatics
Rubin, D. L., Greenspan, H. K., & Brinkley, J. F. (2013) · Springer eBooks · doi:10.1007/978-1-4471-4474-8_9 · Cited by 21
General Purpose (GenP) Bioimage Ensemble of Handcrafted and Learned Features with Data AugmentationPreprint
Nanni, L., Brahnam, S., Ghidoni, S., & Maguolo, G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.08084 · Cited by 25
Automatic CAD System for HEp-2 Cell Image Classification
Ensafi, S., Lu, S., Kassim, A. A., & Tan, C. L. (2014) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2014.573 · Cited by 16
“These discrete words provide the dictionary, which represent the elements of the image and are widely used in object recognition or scene classification [4] [5].”
BackgroundLou, X. P., Zhu, Y., Punithakumar, K., Le, L. H., & Li, B. (2020) · IEEE Access · doi:10.1109/access.2020.3035772 · Cited by 17
“In the training stage, all thoracic CT images that are involved in the experiment are first classified by the bag-of-features method [27], [28].”
MethodA System for Colorectal Tumor Classification in Magnifying Endoscopic NBI Images
Tamaki, T., Yoshimuta, J., Takeda, T., Raytchev, B., Kaneda, K., Yoshida, S., Takemura, Y., & Tanaka, S. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-19309-5_35 · Cited by 16
“While using randomly selected points for computing the descriptors has been reported to show better result [22], we use a regular grid since the texture in NBI images fills the whole image.”
MethodNMF-based multimodal image indexing for querying by visual example
Gonzalez, F. A., Caicedo, J. C., Nasraoui, O., & BenAbdallah, J. (2010) · ACM International Conference on Image and Video Retrieval (CIVR) · doi:10.1145/1816041.1816094 · Cited by 16
“This is consistent with other works in natural image categorization that have reported increased performance when more elements are added to the visual dictionary [20].”
ResultAdaptive Patch Features for Object Class Recognition with Learned Hierarchical Models
Scalzo, F., & Piater, J. (2007) · 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.2007.383371 · Cited by 16
“To specifically evaluate our adaptive patch features (Fig. 5), we also implemented a bag-of-features recognition system similar to Nowak et al. [ 7 ].”
MethodJi, Z., Amirabdollahian, F., Polani, D., & Dautenhahn, K. (2011) · IEEE RO-MAN · doi:10.1109/roman.2011.6005261 · Cited by 16
“images, the BoF method, which is employed in this work, transforms local features into the histogram form, which has been proved effective in applications of image categorisation [13].”
MethodOrgan Detection Using Deep Learning
Shin, H.-C., Orton, M. R., Collins, D. J., Doran, S., & Leach, M. O. (2016) · Elsevier eBooks · doi:10.1016/b978-0-12-802581-9.00007-x · Cited by 17
Hierarchical feature construction for image classification using Genetic Programming
Suganuma, M., Daiki, T., Shirakawa, S., & Nagao, T. (2016) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2016.7844436 · Cited by 13
Spatial co-training for semi-supervised image classification
Hong, Y., & Zhu, W. (2015) · Pattern Recognition Letters · doi:10.1016/j.patrec.2015.06.017 · Cited by 16
Two-stage salient region detection by exploiting multiple priors
Fan, Q., & Qi, C. (2014) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2014.09.003 · Cited by 16
“However, such ability is very important for many computer vision applications, like image retargeting [1], image classification [2], object recognition [3], summarization of a photo collection [4].”
BackgroundSIFTpack: A Compact Representation for Efficient SIFT Matching
Gilinsky, A., & Zelnik‐Manor, L. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.101 · Cited by 16
“This could compromise accuracy, as it was repeatedly shown that dense-sampling yields better results in recognition [9, 14, 19].”
BackgroundCodebook-based electrooculography data analysis towards cognitive activity recognition
Łagodziński, P., Shirahama, K., & Grzegorzek, M. (2017) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2017.10.026 · Cited by 17
Bag of Local Convolutional Triplets for Script Identification in Scene Text
Zdenek, J., & Nakayama, H. (2017) · IAPR International Conference on Document Analysis and Recognition (ICDAR) · doi:10.1109/icdar.2017.68 · Cited by 16
“Vectors of responses extracted from a CNN can be treated like any other descriptors and we can cluster them with k-means [27] or use a Gaussian Mixture Model [28] to generate a dictionary of codewords for a BoVW model.”
MethodFine‐grained recognition of maritime vessels and land vehicles by deep feature embedding
Solmaz, B., Gündoğdu, E., Yücesoy, V., Koç, A., & Alatan, A. A. (2018) · IET Computer Vision · doi:10.1049/iet-cvi.2018.5187 · Cited by 16
“Targeting image recognition for generic objects, researchers presented various types of features and descriptor-based approaches [1, 9, 12] and developed effective methods.”
BackgroundSparsity-regularized feature selection for multi-class remote sensing image classification
Chen, T., Zhao, Y., & Guo, Y. (2019) · Neural Computing and Applications · doi:10.1007/s00521-019-04046-7 · Cited by 16
Majumder, R., Khan, S. M., Ahmed, F., Khan, Z., Ngeni, F., Comert, G., Mwakalonge, J., Michalaka, D., & Chowdhury, M. A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2108.01125 · Cited by 35
“…GPUs, various handcrafted features such as scale-invariant features transform (SIFT) ( 27 ) and histogram of oriented gradients (HOG) are designed and combined to build a bag-of-features (BoF) (28 ) for conventional machine learning models and used for practical application of image classification.”
MethodDetection and Localization of Tip-Burn on Large Lettuce Canopies
Franchetti, B. M., & Pirri, F. (2022) · Frontiers in Plant Science · doi:10.3389/fpls.2022.874035 · Cited by 17
A Survey of Landmark Recognition Using the Bag-of-Words Framework
Bhattacharya, P., & Gavrilova, M. L. (2012) · Studies in computational intelligence · doi:10.1007/978-3-642-31745-3_13 · Cited by 15
Random interest regions for object recognition based on texture descriptors and bag of features
Nanni, L., Brahnam, S., & Lumini, A. (2011) · Expert Systems with Applications · doi:10.1016/j.eswa.2011.07.097 · Cited by 15
“Starting from the conclusion of Nowak et al. (2006), where it is shown that random sampling gives equal or better performance than the sophisticated multiscale interest operators that are in common use, we further simplify the idea of SIFT.”
Enhanced Spatial Pyramid Matching Using Log-Polar-Based Image Subdivision and Representation
Zhang, E., & Mayo, M. (2010) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta.2010.46 · Cited by 15
“Secondly, research by Fei-Fei et al. [8] found that dense features work better for scene classification and that random sampling of keypoints work nearly as well as keypoints selected by detectors [ 16 ].”
BackgroundA Closer Look at Context: From Coxels to the Contextual Emergence of Object Saliency
Mairon, R., & Ben‐Shahar, O. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-10602-1_46 · Cited by 15
“A capacity to predict what is salient or not has also spared much computational resources in image classification [39], retrieval [13], object recognition [43] image and video compression [15,50], and served various other applications such as image thumbnailing [34,45], visualization and symmetrization [47,18,42] and…”
BackgroundNested Sparse Quantization for Efficient Feature Coding
Boix, X., Roig, G., Leistner, C., & Gool, L. V. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33709-3_53 · Cited by 17
“In particular, its computational cost increases with the size of the codebook and the number of patches [10].”
BackgroundHassner, T., Filosof, S., Mayzels, V., & Zelnik‐Manor, L. (2016) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2592916 · Cited by 17
Searching for repeated video sequences
Can, T., & Duygulu, P. (2007) · Multimedia Information Retrieval · doi:10.1145/1290082.1290112 · Cited by 15
“These effects are examined deeply in [19].”
DeepSeaNet: A Bio-Detection Network Enabling Species Identification in the Deep Sea Imagery
Liu, A., Liu, Y., Xu, K., Zhao, F., Zhou, Y., & Li, X. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3359350 · Cited by 14
“…further expressed as Loss ( cella , cellb ) = 0 img ij ( k ) L CIoU represents the loss function of the bounding box based on complete-intersection over union (IOU) [33]. α obj = 0 img ij ( k ) L MSE is the loss of the deep-sea imaging environment, expressed by the form of mean square error (MSE).”
BackgroundA Performance Evaluation of Exact and Approximate Match Kernels for Object Recognition
Barbara Caputo, B., & Luo, J. (2010) · ELCVIA Electronic Letters on Computer Vision and Image Analysis · doi:10.5565/rev/elcvia.350 · Cited by 14
Graph-based methods for the automatic annotation and retrieval of art prints
Carneiro, G. (2011) · International Conference on Multimedia Retrieval · doi:10.1145/1991996.1992028 · Cited by 14
“In this work, the image patch location (in the image) and scale ( patch size) are randomly determined [37], and we generate 1000 des criptors per image.”
MethodLarge-scale classification of traffic signs under real-world conditions
Hazelhoff, L., Creusen, I. M., van de Wouw, D. W. J. M., & de With, P. H. N. (2012) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.910490 · Cited by 14
“In literature, numerous descriptor functions are known, including SIFT,5 SURF,6 DAISY7 and HoG,8 where dense sampling outperforms interest point operators, as shown by Nowak et al.9 Since the amount of extracted descriptors is extremely large, a relatively small codebook is formed, aiming at containing the most common…”
BackgroundEmbedding Visual Words into Concept Space for Action and Scene Recognition
Saghafi, B., Farahzadeh, E., Rajan, D., & Sluzek, A. S. (2010) · British Machine Vision Conference (BMVC) · doi:10.5244/c.24.15 · Cited by 14
“For scene recognition, dense features are more discriminative than sparse ones [16].”
BackgroundMultimodal biomedical image indexing and retrieval using descriptive text and global feature mapping
Simpson, M. S., Demner‐Fushman, D., Antani, S., & Thoma, G. R. (2013) · Information Retrieval · doi:10.1007/s10791-013-9235-2 · Cited by 17
“Alternatively, an interest point detector can be used to detect salient local patches within an image (Nowak et al. 2006).”
BackgroundAutomatic Annotation of Scientific Video Material based on Visual Concept Detection
Hentschel, C., Blümel, I., & Sack, H. (2013) · International Conference on Knowledge Management and Knowledge Technologies · doi:10.1145/2494188.2494213 · Cited by 14
Semi-Supervised Clustering via Learnt Codeword Distances
Batra, D., Sukthankar, R., & Chen, T. (2008) · British Machine Vision Conference (BMVC) · doi:10.5244/c.22.90 · Cited by 14
“Following the notation popularized by bag-of-words models [13], we refer to these initial set of cluster centres as “codewords”.”
MethodKim, A., & Eustice, R. M. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6094820 · Cited by 20
“Recently, several bag-of-words saliency metrics have been explored [ 21 ]‐[24].”
BackgroundSketch retrieval via local dense stroke features
Ma, C., Yang, X., Zhang, C., Ruan, X., & Yang, M. (2016) · Image and Vision Computing · doi:10.1016/j.imavis.2015.11.007 · Cited by 13
“Grid-based as well as random sampling methods have also been proposed to locate local features [32,33].”
BackgroundBag of Features Model Using the New Approaches: A Comprehensive Study
Chougrad, H., Zouaki, H., & Alheyane, O. (2016) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2016.070132 · Cited by 18
Johnson, M., & Shotton, J. (2010) · Studies in computational intelligence · doi:10.1007/978-3-642-12848-6_7 · Cited by 15
“Filter-bank responses (derivatives of Gaussians, wavelets, etc.) or invariant descriptors (e.g., SIFT [16]) are computed across a training set, either at sparse interest points [19] or more densely; recent results in [ 22 ] suggest that densely sampling visual words improves categorization performance.”
ResultEarly Alzheimer disease detection with bag-of-visual-words and hybrid fusion on structural MRI
Ahmed, O. B., Benois‐Pineau, J., Amar, C. B., Allard, M., & Catheline, G. (2013) · International Workshop on Content-Based Multimedia Indexing (CBMI) · doi:10.1109/cbmi.2013.6576557 · Cited by 15
“This fact corresponds with the results of the paper [13].”
ResultLonely but attractive: Sparse color salient points for object retrieval and categorization
Stöttinger, J., Hanbury, A., Gevers, T., & Sebe, N. (2009) · 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2009.5204286 · Cited by 13
“In general, the current trend is toward increasing the number of points [26], applying several detectors or combining them [13, 19], or making the salient point distribution as dense as possible [16, 21].”
Automatic food recognition system for diabetic patients
Velvizhy, P., Pavithra, P., & Kannan, A. (2014) · International Conference on Advanced Computing · doi:10.1109/icoac.2014.7229735 · Cited by 13
A correlation based feature representation for first-person activity recognition
Kahani, R., Talebpour, A., & Mahmoudi-Aznaveh, A. (2019) · Multimedia Tools and Applications · doi:10.1007/s11042-019-7429-3 · Cited by 15
“Although the dense sampling increases the computational cost, the results show that it could improve recognition accuracy [14, 35, 55].”
BackgroundSchmid, B. J., & Adhami, R. R. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2859594 · Cited by 13
Evaluation of local features and classifiers in BOW model for image classification
Qu, Y., Wu, S., Liu, H., Xie, Y., & Wang, H. (2012) · Multimedia Tools and Applications · doi:10.1007/s11042-012-1107-z · Cited by 12
Mining the web for visual concepts
Morsillo, N., Pal, C., & Nelson, R. C. (2008) · doi:10.1145/1509212.1509215 · Cited by 13
“We forego the common approach of interest point detection in light of recent evidence that it is not necessarily more effective than random sampling [15].”
Background
Metadata
- Authors
- Nowak, E., Jurie, F., & Triggs, B.
- Affiliations
- Centre National de la Recherche Scientifique (FR)
- Venue
- Lecture Notes in Computer Science
- Publisher
- Springer Berlin Heidelberg
- Published
- 2006
- Volume
- 3954
- Pages
- 490-503
- ISSN
- 0302-9743 (ISSN-L)
- 1611-3349
- OpenAlex
- W2171896402
- Cited by
- 927
- References
- 26
- Citations by year
- 2026: 1 · 2025: 3 · 2024: 7 · 2023: 10 · 2022: 11 · 2021: 17 · 2020: 35 · 2019: 44 · 2018: 55 · 2017: 59 · 2016: 73 · 2015: 77 · 2014: 78 · 2013: 82 · 2012: 56
- FWCI
- 42.67
- Type
- conference paper
- Language
- en
- Open access
- green
- hal.science
- Licence
- other-oa
- www.springer.com/tdm (tdm)
- Locations
- Lecture notes in computer science (publishedVersion)
- HAL (submittedVersion, other-oa)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image Retrieval and Classification Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Keywords
- bag of visual words, content-based image classification, patch sampling, random sampling, image descriptors, codebook size, histogram normalization, texture analysis
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.