Composition systems

“…The sparse FRAME model is a special case of compositional models advocated by S. Geman et al for vision [22]. In particular, it follows the And-Or grammar studied by Zhu and Mumford [65], where the composition of the wavelets forms an And-node, and the perturbation of each selected wavelet and the variation of its coefficient form an Or-node.…”

Section: Related Workmentioning

confidence: 99%

Learning Sparse FRAME Models for Natural Image Patterns

Xie

Zhu

et al. 2014

Int J Comput Vis

View full text Add to dashboard Cite

It is well known that natural images admit sparse representations by redundant dictionaries of basis functions such as Gabor-like wavelets. However, it is still an open question as to what the next layer of representational units above the layer of wavelets should be. We address this fundamental question by proposing a sparse FRAME (Filters, Random field, And Maximum Entropy) model for representing natural image patterns. Our sparse FRAME model is an inhomogeneous generalization of the original FRAME model. It is a non-stationary Markov random field model that reproduces the observed statistical properties of filter responses at a subset of selected locations, scales and orientations. Each sparse FRAME model is intended to represent an object pattern and can be considered a deformable template. The sparse FRAME model can be written as a shared sparse coding model, which motivates us to propose a twostage algorithm for learning the model. The first stage selects the subset of wavelets from the dictionary by a shared matching pursuit algorithm. The second stage then estimates the parameters of the model given the selected wavelets. Our experiments show that the sparse FRAME models are capable of representing a wide variety of object patterns in natural images and that the learned models are useful for object classification.

show abstract

“…Conceptual advantages of deep hierarchies over 'flat architectures' have been discussed in depth in, e.g., the context of matching [17,22,54]. Tsotsos [82] spelled out the NP completeness of unbounded visual search and gives arguments that hierarchical architectures are a promising way to approach that problem.…”

Section: Retinamentioning

confidence: 99%

Using surfaces and surface relations in an Early Cognitive Vision system

Kraft

Mustafa

Popovic

et al. 2015

Machine Vision and Applications

View full text Add to dashboard Cite

We present a deep hierarchical visual system with two parallel hierarchies for edge and surface information. In the two hierarchies, complementary visual information is represented on different levels of granularity together with the associated uncertainties and confidences. At all levels geometric and appearance information is coded explicitly in 2D and 3D allowing to access this information separately and to link between the different levels. We demonstrate the advantages of such hierarchies in three applications covering grasping, view-point independent object representation, and pose estimation.

show abstract

Composition systems

Cited by 77 publications

References 25 publications

Redesigning Systems Thinking

Redesigning Systems Thinking

Learning Sparse FRAME Models for Natural Image Patterns

Using surfaces and surface relations in an Early Cognitive Vision system

Contact Info

Product

Resources

About