1 / 18

Articulated People Detection and Pose Estimation: Reshaping the Future

Articulated People Detection and Pose Estimation: Reshaping the Future. Leonid Pishchulin Arjun Jain Mykhaylo Andriluka Thorsten Thorm¨ahlen Bernt Schiele Max Planck Institute for Informatics, Saarbr¨ucken , Germany. OUTLINE. Introduction Generation of novel training examples

homer
Download Presentation

Articulated People Detection and Pose Estimation: Reshaping the Future

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. Articulated People Detection and Pose Estimation: Reshaping the Future Leonid PishchulinArjun JainMykhayloAndriluka Thorsten Thorm¨ahlenBerntSchiele Max Planck Institute for Informatics, Saarbr¨ucken, Germany

  2. OUTLINE • Introduction • Generation of novel training examples • Articulated people detection • Articulated pose estimation • Articulated pose estimation “in the wild” • Conclusion

  3. Introduction Recent progress in people detection and articulated pose estimation may be contributed to two key factors. • Discriminative learning allows to learn powerful models on a large training corpora • robust image features enable to deal with image clutter, occlusions and appearance variation

  4. Introduction

  5. Generation of novel training examples

  6. Generation of novel training examples

  7. Articulated people detection We use the deformable part model (DPM) [11] and evaluate its performance on the “Image Parsing” dataset [25]. For training we use training sets from the publicly available datasets: • DPM-VOC PASCAL VOC 2009 (VOC) [10] • DPM-IP “Image Parsing”(IP) [25] • DPM-LSP “Leeds Sports Poses” (LSP)dataset [19] • DPM-IP-R and DPM-IP-AR [10] M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman. The PASCAL visual object classes (VOC) challenge.IJCV’10. [11] P. F. Felzenszwalb, R. B. Girshick, D. McAllester, and D. Ramanan.Objectdetection with discriminatively trained part-based models.PAMI’10. [19] S. Johnson and M. Everingham. Clustered pose and nonlinear appearance models for human pose estimation. In BMVC’10. [25] D. Ramanan. Learning to parse images of articulated objects. In NIPS’06.

  8. Articulated people detection

  9. Articulated people detection

  10. Articulated people detection

  11. Articulated pose estimation • Proposes a new joint model for body pose estimation combining pictorial structures [12,14]model with DPM

  12. Articulated pose estimation

  13. Articulated pose estimation

  14. Articulated pose estimation “in the wild” We define a new dataset based on the LSP by using the publicly available original non-cropped images. This dataset, in the following denoted as “multi-scale LSP”

  15. Articulated pose estimation “in the wild”

  16. Articulated pose estimation “in the wild”

  17. Conclusion • Propose a novel method for automatic generation of training examples • Evaluate our data generation method for articulated people detection and pose estimation and show that we significantly improve the performance • Propose a joint model

  18. Thank you for listening

More Related