Poselets: Body Part Detectors Trained Using 3D Human Pose Annotations

We address the classic problems of detection, segmenta- tion and pose estimation of people in images with a novel definition of a part, a poselet. We postulate two criteria (1) It should be easy to find a poselet given an input image (2) it should be easy to localize the 3D configuration of the person conditioned on the detection of a poselet. To permit this we have built a new dataset, H3D, of annotations of humans in 2D photographs with 3D joint information, in- ferred using anthropometric constraints. This enables us to implement a data-driven search procedure for finding pose- lets that are tightly clustered in both 3D joint configuration space as well as 2D image appearance. The algorithm dis- covers poselets that correspond to frontal and profile faces, pedestrians, head and shoulder views, among others. Each poselet provides examples for training a linear SVM classifier which can then be run over the image in a multiscale scanning mode. The outputs of these posel...

• Type: Conference
• Year: 2009
• Where: ICCV
• Authors: Lubomir Bourdev, Jitendra Malik
• Keywords: Body Part Detectors, Computer Vision, Human Pose Annotations, ICCV 2009