Computer Science Graduate Seminar: 3D Shape Analysis based on Feature Curve Networks
Tuesday, February 12, 2019, 3:00pm
Location: Building E3, Seminar room 118, Ahornstr. 55
Speaker: Anne Gehre M.Sc. (i8)
For high-level analysis of 3D shapes, we require an abstract representation of geometric data. Typically, this is achieved by developing descriptors on a local pointwise level or globally on the entire shape. While point based descriptors can be very sensitive to local changes of the shape (e.g. noise), global descriptors tend to be too coarse.
Feature curves trace out salient creases and crests of 3D geometric data. They provide an abstract representation of salient parts of the geometry and contain topological and global structural information about the shape as well as geometric details. However, automatically computed feature curve networks on raw data can have various defects such as noise, fragmentation, or missing data.
In this talk we present methods to asses the different abstraction levels of a feature curve network. This is vital in order to obtain a meaningful set of feature curves. For this we combine local (feature curve strength, length, parallelism, etc.) and global (density and symmetry) saliency measures and obtain structural reoccurrence information from a raw, potentially noisy input data. In the context of high-level shape analysis we discuss how meaningful feature curve networks have a highly positive impact on complex tasks such as the discovery of inter-surface maps.
The computer science lecturers invite interested people to join.