1 / 19

Object Detection by Matching

Object Detection by Matching. Longin Jan Latecki. ?. Contour-based object detection. Database shapes: ….. . Easy for Human Eyes. Everybody can find the swan in these images. Recall humans can draw a swan. We always draw edges.

raven
Download Presentation

Object Detection by Matching

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. Object Detection by Matching Longin Jan Latecki

  2. ? Contour-based object detection Database shapes: …..

  3. Easy for Human Eyes • Everybody can find the swan in these images.

  4. Recall humans can draw a swan • We always draw edges.

  5. Computer can also capture edges

  6. Humans can detect shapes given only edges

  7. Problem: separating noise • Too much noise, and the computer can’t tell which edges belong to object of interest.

  8. Object recognition process: Source: 2D image of a 3D object Object Segmentation Contour Extraction Contour Cleaning, e.g., Evolution Contour Segmentation Matching to database shapes

  9. Object detection as matching database shapes to image edge segments Database shapes: ….. ? matching Edge detection Edge linking Contour grouping

  10. Main challenges 1 . The contour of the desired object is typically fragmented over several pieces. 2 . Part of the true contour of the target object may be wrongly connected to part of a background contour resulting in a single edge fragment

  11. Problem formulation ? How to find the true contours of the target shape in the edge image? Key idea: Given a minimal required coverage of the model contour, we want to select non overlapping model fragments that maximize the configuration similarity to the corresponding image fragments.

  12. All relevant edge fragments are mapped to their corresponding model fragments. Key idea: Build an association graph. Find maximum weight subgraph

  13. Construction of Affinity Matrix The weighted affinity graph is denoted as G = (V, A). Each vertex of the graph corresponds to a partial match The affinity between node i and node j is based on their shape similarity.

  14. Construction of Affinity Matrix High affinity: Low affinity:

  15. Problem with Affinity Matrix Wrong matches may also have high affinity:

  16. We use model and image location constraints to sparsify the Affinity Matrix

  17. Maximal Cliques in a Weighted Graph A maximal clique is a subset of V with maximal average affinity between all pairs of its vertices. [ M. Pavan and M. Pelillo. PAMI 2007] In this example, the maximal clique has 4 nodes selected from over 500 nodes. Therefore, most clustering based approach may not succeed.

  18. Computing Maximum Weight Subgraphs Indicator = selected maximal clique of vertices of V. In order to solve this combinational problem, we relax it to A vertex is selected as belonging to a MWS iff Each MWS corresponds to a local maximum of: Each local solution is not a final solution but a detection hypothesis. Tianyang Ma and Longin Jan Latecki. From Partial Shape Matching through Local Deformation to Robust Global Shape Similarity for Object Detection. CVPR 2011.

  19. Object detection examples

More Related