In the first part, we present an efficient curve alignment algorithm derived from the congealing framework that is effective on many synthetic and real data sets.
We show that using the byproducts of joint alignment, the aligned data and transformation parameters, can dramatically improve classification performance.
The chordiogram descriptor applies holistic properties of the shape and also proven suitable for object detection and digit recognition mechanisms. This thesis describes an approach for object recognition using the chordiogram shape-based descriptor.
The employed shape descriptor chordiogram is based on geometric relationships of chords generated from the pairs of boundary points of an object.
Semantic labeling is an important mid-level vision task for grouping and organizing image regions into coherent parts.
Labeling these regions allows us to better understand the scene itself as well as properties of the objects in the scene, such as their parts, location, and interaction within the scene.Typical approaches for this task include the conditional random field (CRF), which is well-suited to modeling local interactions among adjacent image regions.However the CRF is limited in dealing with complex, global (long-range) interactions between regions in an image, and between frames in a video.In spite of such excellent properties, chordiogram is not scale-invariant.To this end, we propose scale invariant chordiogram descriptors and intend to achieve a similar performance before and after applying scale invariance.This thesis presents approaches to modeling long-range interactions within images and videos, for use in semantic labeling.In order to model these long-range interactions, we incorporate priors based on the restricted Boltzmann machine (RBM).Theses and dissertations represent a wealth of scholarly and artistic content created by masters and doctoral students in the degree-seeking process.Some ETDs in this collection are restricted to use by the UNT community.Second, they require hand-picking appropriate feature representations for each data set, which may be time-consuming and ineffective, or outside the domain of expertise for practitioners.In this thesis we introduce alignment models that address both shortcomings.