Probabilistic visual learning for object representation

Moghaddam, B.; Pentland, Alex

doi:10.1109/34.598227

Cited by 1,219 publications

(710 citation statements)

References 35 publications

Supporting

Mentioning

703

Contrasting

Unclassified

Order By: Relevance

“…the Gaussian mixture model allows the soft partition of data points in proportion to the responsibility defined by the posterior probability which indicates the relative probability that the data was derived from each class. Moghaddam and Pentland (1996), for example, presented a method that combines the principal component analysis with a Gaussian mixture model for object recognition and detection. As the dimensionality of object images is in general very high in the input space, the method first finds a principal sub-space, and then the Gaussian mixture model is applied to estimate the multimodal densities within the principal sub-space.…”

Section: Gaussian Mixture Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised visual learning of three-dimensional objects using a modular network architecture

1999

View full text Add to dashboard Cite

This paper presents a modular network architecture that learns to cluster multiple views of multiple three-dimensional (3D) objects. The proposed network model is based on a mixture of non-linear autoencoders, which compete to encode multiple views of each 3D object. The main advantage of using a mixture of autoencoders is that it can capture multiple non-linear sub-spaces, rather than multiple centers for describing complex shapes of the view distributions. The unsupervised training algorithm is formulated within a maximum-likelihood estimation framework. The performance of the modular network model is evaluated through experiments using synthetic 3D wire-frame objects and gray-level images of real 3D objects. It is shown that the performance of the modular network model is superior to the performance of the conventional clustering algorithms, such as the K-means algorithm and the Gaussian mixture model. ᭧

show abstract

Section: Gaussian Mixture Modelsmentioning

confidence: 99%

“…Poggio & Edelman, 1990;Murase & Nayar, 1995;Moghaddam & Pentland, 1996). This approach contrasts with a conventional approach that constructs structural descriptions of objects using 3D volumetric primitives in the object-centered coordinate frame (e.g.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised visual learning of three-dimensional objects using a modular network architecture

1999

View full text Add to dashboard Cite

show abstract

“…The subsequent swimmer detection is based on the approach of Shechtman 16 who also introduced the self-similarity descriptors. The underlying probabilistic matching approach was suggested by Moghaddam et al 17 Furthermore, we compare the performance of the self-similarity features to several alternative feature descriptors such as SIFT, 18 Geometric Blur 19 and HOG. 20 Finally, three-dimensional joint positions are estimated by applying the method of Taylor.…”

Section: Related Workmentioning

confidence: 99%

Automatic pose initialization of swimmers in videos

Ries¹,

Lienhart²

2010

SPIE Proceedings

View full text Add to dashboard Cite

We propose an approach to the task of automatic pose initialization of swimmers in videos. Thus, our goal is to detect a swimmer inside a target video and assign an estimated position to her/his body parts. We first apply a non-skin-color filter to reduce the search space inside each target frame. We then match previously devised template sequences of Gaussian feature descriptors against sequences of feature vectors which are computed within the remaining image regions. Finally, relative average joint positions from annotated images featuring the key pose are assigned to the detection result and three-dimensional joint positions are estimated. We present detection results for test videos of three different swim strokes and examine the performance of four types of feature descriptors.

show abstract

“…Sparse structuring of statistical dependency explains the empirical success of "parts-based" methods for face detection and face recognition [1][2] [3][4] [5]. Such parts-based methods concentrate modeling power on localized regions, in which dependencies tend to be strong and use weaker models over larger areas in which dependencies tend to be less significant.…”

Section: Introductionmentioning

confidence: 99%

Learning Statistical Structure for Object Detection

Schneiderman

2003

Computer Analysis of Images and Patterns

View full text Add to dashboard Cite

Abstract. Many classes of images exhibit sparse structuring of statistical dependency. Each variable has strong statistical dependency with a small number of other variables and negligible dependency with the remaining ones. Such structuring makes it possible to construct a powerful classifier by only representing the stronger dependencies among the variables. In particular, a seminaïve Bayes classifier compactly represents sparseness. A semi-naïve Bayes classifier decomposes the input variables into subsets and represents statistical dependency within each subset, while treating the subsets as statistically independent. However, learning the structure of a semi-naïve Bayes classifier is known to be NP complete. The high dimensionality of images makes statistical structure learning especially challenging. This paper describes an algorithm that searches for the structure of a semi-naïve Bayes classifier in this large space of possible structures. The algorithm seeks to optimize two cost functions: a localized error in the log-likelihood ratio function to restrict the structure and a global classification error to choose the final structure. We use this approach to train detectors for several objects including faces, eyes, ears, telephones, push-carts, and door-handles. These detectors perform robustly with a high detection rate and low false alarm rate in unconstrained settings over a wide range of variation in background scenery and lighting.

show abstract

Probabilistic visual learning for object representation

Cited by 1,219 publications

References 35 publications

Unsupervised visual learning of three-dimensional objects using a modular network architecture

Unsupervised visual learning of three-dimensional objects using a modular network architecture

Automatic pose initialization of swimmers in videos

Learning Statistical Structure for Object Detection

Contact Info

Product

Resources

About