Intel

Courses

Structure-Aware Shape Processing

Monday, 11 August 2:00 PM - 5:15 PM | Vancouver Convention Centre, West Building, Rooms 211-214

Shape structure is based on the arrangement and relationships among shape parts. Structure-aware shape processing goes beyond local geometry and low-level processing to analyze and process shapes at a high level. It focuses more on the global inter- and intra-semantic relations among the parts of shape rather than on their local geometric details.

With recent developments in easy shape acquisition, access to vast repositories of 3D models, and simple-to-use desktop fabrication possibilities, the study of structure in shapes has become a central research topic in shape analysis, editing, and modelling. A whole new line of structure-aware shape processing algorithms has emerged. Their operation is based on an attempt to understand such structure in shapes and, eventually, link object functions to shape geometry. The algorithms broadly consist of two key phases: an analysis phase, which extracts structural information from input data; and a (smart) processing phase, which utilizes the extracted information for exploration, editing, and synthesis of novel shapes.

Course Schedule

2 pm
Introduction to Geometric Structure
Mitra

(Analysis) Extracting Structures
2:15 pm
A. Analysis of Individual Models
Wand

2:45 pm
B. Analysis of Shape Collections
Huang

3:20 pm
C. Encoding Structural Hierarchy
Zhao

3:40 pm
Break

(Application) Manipulating Structures
3:50 pm
A. Modeling as Structural Variations
Zhao

4:10 pm
B. Structure-Guided Design
Mitra

4:30 pm
C. Organization and Exploration of Shape Collections
Kim

5 pm
Conclusion and Open Problems
All

Level

Intermediate

Prerequisites

Basic background in computer graphics and particularly geometric modelling and processing. Working knowledge of data analysis and machine learning recommended but not required.

Intended Audience

Graduate students, practitioners, and researchers interested in shape analysis and, in particular, massive shape collections.

Instructor(s)

Niloy Mitra
University College London, King Abdullah University of Science and Technology

Michael Wand
Universiteit Utrecht

Hao Zhang
Simon Fraser University

Daniel Cohen-Or
Tel Aviv University

Vadimir Kim
Stanford University

Qi-Xing Huang
Stanford University