|Ph.D Student||Leifman George|
|Subject||Similarity for Shape Analysis|
|Department||Department of Electrical Engineering||Supervisor||Professor Ayellet Tal|
|Full Thesis text|
In this thesis we discuss two shape analysis problems, which are based on shape similarity: saliency and colorization. First, we propose an algorithm for saliency detection of triangulated meshes and of point clouds. While saliency detection in images has been extensively studied, relatively few papers have addressed objects in 3D. Our algorithm for mesh saliency detection looks for regions that are distinct both locally and globally and accounts for the distance to the foci of attention. It is also shown how this algorithm can be adopted to saliency detection in point clouds. Saliency can be utilized in many applications; we explore viewpoint selection; the most informative views are those that collectively provide the most descriptive presentation of the surface. The second problem discussed in the thesis is mesh colorization. Traditionally, colorization refers to the process of adding color to black & white images or videos. Our goal is to extend it to surfaces in three dimensions. This is important for applications in which the colors of an object need to be restored and no relevant image exists for texturing it. In such cases, if the user could easily and quickly colorize the mesh with a few brush strokes, texturing would be avoided. First we propose such a colorization algorithm, where the colorization is formulated as a constrained quadratic optimization problem. Special care is taken to avoid color bleeding between regions, through the definition of a new direction field on meshes.
Then, we focus on surfaces with patterns and propose an algorithm for adding colors to these surfaces. The user needs only to scribble a few color strokes on one instance of each pattern, and the system proceeds to automatically colorize the whole surface.
For this scheme to work, we address not only the problem of colorization, but also the problem of pattern detection on surfaces.