Filters








16,351 Hits in 4.3 sec

Patch-based Object Recognition Using Discriminatively Trained Gaussian Mixtures

A. Hegerath, T. Deselaers, H. Ney
2006 Procedings of the British Machine Vision Conference 2006  
We present an approach using Gaussian mixture models for part-based object recognition where spatial relationships of the parts are explicitly modeled and parameters of the generative model are tuned discriminatively  ...  These extensions lead to great improvements of the classification accuracy.  ...  Gaussian Mixture Models Gaussian mixture models are a generative model: for each object class a class-dependent mixture p(x | k) is used.  ... 
doi:10.5244/c.20.54 dblp:conf/bmvc/HegerathDN06 fatcat:dpvurruosfdchh2odroa4ot5f4

An Effective Modeling for Face Recognition System: LDA and GMM based Approach

Aditi Mandloi, Priyanka Gupta
2017 International Journal of Computer Applications  
In this work we presented a novel Face Recognition feature Extraction Mode based on the combination of Linear Discriminant Analysis (LDA) and Gaussian Mixture Model (GMM).  ...  For the betterment of the feature classification a KNN classifier is used.  ...  To this end, we represent image set as a Gaussian Mixture Model (GMM) comprising a number of Gaussian components with prior probabilities and seek to discriminate Gaussian components from different classes  ... 
doi:10.5120/ijca2017915889 fatcat:irlohfxeevhdtle42p66msep6a

Face Recognition System Using: LDA and GMM based Approach

Aditi Mandloi, Priyanka Gupta
2017 International Journal of Computer Applications  
In this work we presented a novel Face Recognition feature Extraction Mode based on the combination of Linear Discriminant Analysis (LDA) and Gaussian Mixture Model (GMM).  ...  For the betterment of the feature classification a KNN classifier is used.  ...  To this end, we represent image set as a Gaussian Mixture Model (GMM) comprising a number of Gaussian components with prior probabilities and seek to discriminate Gaussian components from different classes  ... 
doi:10.5120/ijca2017915888 fatcat:4o7zzg556na77jhixwnedu25ze

Research on Saliency Prior Based Image Processing Algorithm

Zhouping Yin, Hongmei Zhang
2014 Journal of Multimedia  
Current image processing model still need large amount of training data to tune processing model and can't process large images effectively.  ...  Therefore, this paper researched saliency prior based image processing model, present the Gaussian mixture process and design the feature point based classifier, and then evaluate the model by supervised  ...  Estimated the test sample frames image distribution for each channel by Gaussian mixture models 3.  ... 
doi:10.4304/jmm.9.2.294-301 fatcat:engsho3uozam7eqxvkdrt6wsc4

Object recognition using proportion-based prior information: Application to fisheries acoustics

R. Lefort, R. Fablet, J.-M. Boucher
2011 Pattern Recognition Letters  
The main contribution of this work is the development of learning methods for training datasets consisting of groups of objects with known relative class priors.  ...  Research highlights ► Weakly supervised learning deals with prior annotation of objects in images. ► Classification model must be assessed by using probabilities. ► Reported results promote discriminative  ...  Conclusion The development of reliable methods for object classification and recognition in images is an active area of research.  ... 
doi:10.1016/j.patrec.2010.10.001 fatcat:cicrhinhkzfqvlpvvb6bcs2x3u

A Trainable Object-Detection Method Using Equivalent Retinotopical Sampling and Fisher Kernel [chapter]

Hirotaka Niitsuma
2003 Lecture Notes in Computer Science  
The second method is an extension of discriminant function trained by SVMs for object recognition in an image.  ...  The discriminant function is formulated as an analytical function of the object position and the object size in an image. The extension is introducing ERS to SVMs.  ...  In this paper, an extension of a discriminant function trained by SVM for object recognition in an image is suggested.  ... 
doi:10.1007/978-3-540-45226-3_22 fatcat:u7iixlj3ybcr5glwqlslvx76e4

Selection of scale-invariant parts for object class recognition

Dorko, Schmid
2003 Proceedings Ninth IEEE International Conference on Computer Vision  
A classifier is then learned for each of these parts, and feature selection is used to determine the most discriminative ones.  ...  This approach allows robust part detection, and it is invariant under scale changes-that is, neither the training images nor the test images have to be normalized.  ...  Agarwal for providing the car images.  ... 
doi:10.1109/iccv.2003.1238407 dblp:conf/iccv/DorkoS03 fatcat:6owlakhjhfhcvp2trjeyordj4a

Hierarchical Gaussianization for image classification

Xi Zhou, Na Cui, Zhen Li, Feng Liang, Thomas S Huang
2009 2009 IEEE 12th International Conference on Computer Vision  
After such a hierarchical Gaussianization, each image is represented by a Gaussian mixture model (GMM) for its appearance, and several Gaussian maps for its spatial layout.  ...  First, we model the feature vectors, from the whole corpus, from each image and at each individual patch, in a Bayesian hierarchical framework using mixtures of Gaussians.  ...  Unlike in scene/object recognition task, the spatial layout does not show much benefit in performance for face recognition.  ... 
doi:10.1109/iccv.2009.5459435 dblp:conf/iccv/ZhouCLLH09 fatcat:oqf2dg3s2vakfcrmpkyrlsmylq

Log-Linear Mixtures for Object Class Recognition

Tobias Weyand, Thomas Deselaers, Hermann Ney
2009 Procedings of the British Machine Vision Conference 2009  
We present log-linear mixture models as a fully discriminative approach to object category recognition which can, analogously to kernelised models, represent non-linear decision boundaries.  ...  We show that this model is the discriminative counterpart to Gaussian mixtures and that either one can be transformed into the respective other.  ...  For each class, the task is to detect whether an object of the class is present in an image. For each task, about 2600 training images and about 2700 testing images are given.  ... 
doi:10.5244/c.23.30 dblp:conf/bmvc/WeyandDN09 fatcat:trqdo7f7tfeplkylpsppad3sxu

Object classification by fusing SVMs and Gaussian mixtures

Thomas Deselaers, Georg Heigold, Hermann Ney
2010 Pattern Recognition  
We present a new technique that employs support vector machines (SVMs) and Gaussian mixture densities (GMDs) to create a generative/discriminative object classification technique using local image features  ...  In the past, several approaches to fuse the advantages of generative and discriminative approaches were presented, often leading to improved robustness and recognition accuracy.  ...  Object Classification Using Local Features with Gaussian Mixture Densities Gaussian mixture models are a generative model: for each object class a class-dependent mixture p(x|k) is used.  ... 
doi:10.1016/j.patcog.2010.02.002 fatcat:t4g4nn6nr5bazcyzc5ypl5gkza

Combined Generative-Discriminative Learning for Object Recognition using Local Image Descriptors

Abhikesh Nag, David J. Miller, Andrew P. Brown, Kevin J. Sullivan
2007 Machine Learning for Signal Processing  
We present a system for scale and affine invariant recognition of vehicular objects in video sequences. We use local descriptors (SIFT keypoints) from image frames to model the object.  ...  This presents certain challenges for classification techniques, which generally require use of the same set of features for every instance of an object to be classified.  ...  Keypoint-Hybrid Generative-Discriminative system Figure 2 . 1 . 21 Location and orientation of SIFT keypoints introduces and motivates the use of Gaussian mixture models (GMM) in object recognition  ... 
doi:10.1109/mlsp.2007.4414333 fatcat:bjp7eo6afjcwbkp3o6uoi4ia5y

A cluster-based statistical model for object detection

T.D. Rikert, M.J. Jones, P. Viola
1999 Proceedings of the Seventh IEEE International Conference on Computer Vision  
The distribution of feature v e ctors for a set of training images of an object class is estimated by clustering the data and then forming a mixture o f gaussian model.  ...  This paper presents an approach to object detection which is based on recent work in statistical models for texture synthesis and recognition 7, 4, 23, 17 .  ...  Ruell and Gene Preble for collecting the face database used in this research.  ... 
doi:10.1109/iccv.1999.790386 dblp:conf/iccv/RikertJV99 fatcat:yowwrefehvfcreuym3n5bk7oem

A generative/discriminative learning algorithm for image classification

Yi Li, L.O. Shapiro, J.A. Bilmes
2005 Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1  
We have developed a two-phase generative / discriminative learning procedure for the recognition of classes of objects and concepts in outdoor scenes.  ...  Our method uses both multiple types of object features and context within the image.  ...  Note that the mixture for object class o is trained with all regions of all images that contain o, but these images also contain many other regions from other object classes.  ... 
doi:10.1109/iccv.2005.7 dblp:conf/iccv/LiSB05 fatcat:g67sugjn4bcwlilurl66wma6xu

Mixture Density Generative Adversarial Networks

Hamid Eghbal-zadeh, Werner Zellinger, Gerhard Widmer
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
This is achieved by positioning Gaussian density functions in the corners of a simplex, using the resulting Gaussian mixture as a likelihood function over discriminator embeddings, and formulating an objective  ...  function for GAN training that is based on these likelihoods.  ...  We also gratefully acknowledge the support of NVIDIA Corporation with the donation of a Titan X GPU used for this research.  ... 
doi:10.1109/cvpr.2019.00597 dblp:conf/cvpr/Eghbal-zadehZW19 fatcat:g2awz7zqz5hfbbvnwso3f2ggcu

Representing Videos as Discriminative Sub-graphs for Action Recognition [article]

Dong Li and Zhaofan Qiu and Yingwei Pan and Ting Yao and Houqiang Li and Tao Mei
2022 arXiv   pre-print
for recognition.  ...  For each action category, we execute online clustering to decompose the graph into sub-graphs on each scale through learning Gaussian Mixture Layer and select the discriminative sub-graphs as action prototypes  ...  For one specific scale, we utilize Gaussian Mixture Layer to learn the discriminative sub-graphs/action prototypes for action recognition.  ... 
arXiv:2201.04027v1 fatcat:4xcaxj77brhyvhwfkgp5pg2gqq
« Previous Showing results 1 — 15 out of 16,351 results