Factorized Local Appearance Models
نویسندگان
چکیده
We propose a novel local appearance modeling method for object detection and recognition in cluttered scenes. The approach is based on the joint distribution of local feature vectors at multiple salient points and factorization with Independent Component Analysis (ICA). The resulting non-parametric densities are simple multiplicative histograms. This leads to computationally tractable joint probability densities which can model high-order dependencies. Testing and evaluation shows that the factorized density model with spatial encoding improves modeling accuracy and outperforms global appearance models in image/object retrieval. Furthermore, experiments in detection of substantially occluded objects in cluttered scenes have demonstrated promising results .
منابع مشابه
Learning Patch-based Structural Element Models with Hierarchical Palettes Abstract Learning Patch-based Structural Element Models with Hierarchical Palettes
Learning Patch-Based Structural Element Models With Hierarchical Palettes Jeroen Chua Master of Applied Science Graduate Department of Electrical and Computer Engineering University of Toronto 2012 Image patches can be factorized into ‘shapelets’ that describe segmentation patterns, and palettes that describe how to paint the segments. This allows a flexible factorization of local shape (segmen...
متن کاملICA-based probabilistic local appearance models
This paper proposes a novel image modeling scheme for object detection and localization. Object appearance is modeled by the joint distribution of k-tuple salient point feature vectors which are factorized component-wise after an independent component analysis (ICA). Also, we propose a distance-sensitive histograming technique for capturing spatial dependencies. The advantages over existing tec...
متن کاملModeling Human Motion Using Manifold Learning and Factorized Generative Models
OF THE DISSERTATION Modeling Human Motion Using Manifold Learning and Factorized Generative Models by Chan-Su Lee Dissertation Director: Ahmed Elgammal Modeling the dynamic shape and appearance of articulated moving objects is essential for human motion analysis, tracking, synthesis, and other computer vision problems. Modeling the shape and appearance of human motion is challenging due to the ...
متن کاملA Comparative Study of the Concept of "Privacy" in the House of Islamic Countries in the Middle East (Case Study: Houses of Isfahan, Sanaa, Damascus)
One of the most important dimensions in the design of Islamic buildings is the creation of privacy. The word ‘privacy’ means ‘environment’, referring to a place where its support and protecting is obligatory and as a whole, it is a barrier to avoid others’ assault. It is of utmost importance in traditional houses especially in Islamic countries the houses separate the family privacy from the st...
متن کاملFactorization for Probabilistic Local Appearance Models
We propose a novel local appearance modeling method for object detection and recognition in cluttered scenes. The approach is based on the joint distribution of local feature vectors at multiple salient points and factorization with Independent Component Analysis (ICA). The resulting non-parametric densities are simple multiplicative histograms. This leads to computationally tractable joint pro...
متن کامل