1 / 26

Contributions

Contributions. A people dataset of 8035 images. 2. 1. Three layer attribute classification framework using poselets. H3D (Humans in 3D). Pascal VOC 2010 (trn+val). 9 different attributes in total. At least 2 attributes per image Agreement of 4 out of 5. TRN: 2003. VAL: 2011.

hisano
Download Presentation

Contributions

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. Contributions A people dataset of 8035 images. 2 1 Three layer attribute classification framework using poselets.

  2. H3D (Humans in 3D) Pascal VOC 2010 (trn+val) 9 different attributes in total. At least 2 attributes per image Agreement of 4 out of 5 TRN: 2003 VAL: 2011 TEST: 4022 People Dataset 1 8035 images

  3. Attribute classification using poselets 2

  4. What is a Poselet ? Poselets capture part of the pose from a given viewpoint [Bourdev & Malik, ICCV09]

  5. Poselets Examples may differ visually but have common semantics [Bourdev & Malik, ICCV09]

  6. Poselets But how are we going to create training examples of poselets?

  7. How do we train a poselet for a given pose configuration?

  8. Finding correspondences at training time Given part of a human pose How do we find a similar pose configuration in the training set?

  9. Finding correspondences at training time Left Shoulder Left Hip We use keypoints to annotate the joints, eyes, nose, etc. of people

  10. Finding correspondences at training time Residual Error

  11. Training poselet classifiers Residual Error: 0.15 0.20 0.10 0.35 0.15 0.85 Given a seed patch Find the closest patch for every other person Sort them by residual error Threshold them

  12. Training poselet classifiers Given a seed patch Find the closest patch for every other person Sort them by residual error Threshold them Use them as positive training examples to train a linear SVM with HOG features

  13. Which poselets should we train? • Choose thousands of random windows, generate poselet candidates, train linear SVMs • Select a small set of poselets that are: • Individually effective • Complementary

  14. Some Poselets

  15. Attribute classification using poselets 2

  16. Features • HOGs at two levels (~2K-4K features) • 16 x 16 • 32 x 32 • Color Histograms in H,S,B (30 features) • 10 bins for H, S and B • Skin classifier output (3 features) • GMM with 5 components • Fraction of skin pixels • hands-skin, legs-skin, neck skin

  17. Poselet-level Attribute Classifiers

  18. Person-level Attribute Classifiers

  19. Context-level Attribute Classifiers

  20. Results

  21. Results

  22. Results

  23. Results

  24. Gender Classification Results

  25. Thanks

More Related