Presentation is loading. Please wait.

Presentation is loading. Please wait.

1 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Michael Bronstein Computational metric geometry: an old new tool in image.

Similar presentations


Presentation on theme: "1 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Michael Bronstein Computational metric geometry: an old new tool in image."— Presentation transcript:

1 1 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Michael Bronstein Computational metric geometry: an old new tool in image sciences

2 2 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity

3 3 retrievalcategorization tracking detection/recognition restorationalignment Similarity

4 4 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Raffaello Santi, School of Athens, Vatican

5 5 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Shape similarity and correspondence Metric space Correspondence Correspondence quality = metric distortion Similarity Gromov-Hausdorff distance = Minimum possible correspondence distortion

6 6 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Invariance RigidInelasticTopologyScaleElastic Choice of the metric prescribes the invariance

7 7 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Non-rigid shape analysis and synthesis BBK CorrespondenceMorphing Retrieval

8 8 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Self-similarity and symmetry Permutation Raviv & BBK 2007

9 9 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Metric learning Data spaceEmbedding space Min distortion on training set of examples with known

10 10 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Video copy detection Luke vs Vader – Starwars classic Lightsaber Star Wars DVD copyStar Wars pirated copy BBK 2010

11 11 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Challenges Theoretical Approximate symmetry notion: group-like structure Comparing data from different spaces Computational Efficient solution of minimum distortion correspondence problems (Gromov-Hausdorff distance) Efficient algorithms for embedding into interesting metric spaces Applications Problems that can be formulated in terms of metric geometry


Download ppt "1 Michael Bronstein Shapes as metric spaces: deformation-invariant similarity Michael Bronstein Computational metric geometry: an old new tool in image."

Similar presentations


Ads by Google