1 / 11

Cutting-Plane Training of Non-associative Markov Network for 3D Point Cloud Segmentation

Cutting-Plane Training of Non-associative Markov Network for 3D Point Cloud Segmentation. Roman Shapovalov , Alexander Velizhev Lomonosov Moscow State University. Hangzhou, May 18, 2011. Semantic segmentation of point clouds. LIDAR point cloud without color information

simon-moreno
Download Presentation

Cutting-Plane Training of Non-associative Markov Network for 3D Point Cloud Segmentation

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Cutting-Plane Training of Non-associative Markov Network for 3D Point Cloud Segmentation Roman Shapovalov, Alexander Velizhev Lomonosov Moscow State University Hangzhou, May 18, 2011

  2. Semantic segmentation of point clouds • LIDAR point cloud without color information • Class label for each point

  3. System workflow featurecomputation CRFinference segmentation graphconstruction

  4. Non-associative CRF node features edge features point labels • Associative CRF: • Our model: no such constraints! [Shapovalov et al., 2010]

  5. CRF training • parametric model • parameters need to be learned! CRFinference

  6. Structured learning [Anguelov et al., 2005; and a lot more] • Linear model: • CRF negative energy: • Find such that …

  7. Structured loss • Define • Define structured loss, for example: • Find such that …

  8. Cutting-plane training • A lot of constraints (Kn) • Maintain a working set • Add iteratively the most violated one: • Polynomial complexity • SVMstruct implementation [Joachims, 2009]

  9. Analysis • Advantages: • more flexible model • accounts for class imbalance • allows kernelization • Disadvantages: • really slow (esp. with kernels) • learns small/underrepresented classes badly

  10. Results [Munoz et al., 2009] Our method

  11. Results Aerial Road

More Related