We address the problem of detecting people in natural scenes using a part approach based on poselets. We propose a bootstrapping method that allows us to collect millions of weakly labeled examples for each poselet type. We use these examples to train a Convolutional Neural Net to discriminate different poselet types and separate them from the background class. We then use the trained CNN as a way to represent poselet patches with a Pose Discriminative Feature (PDF) vector -- a compact 256-dimensional feature vector that is effective at discriminating pose from appearance. We train the poselet model on top of PDF features and combine them with object-level CNNs for detection and bounding box prediction. The resulting model leads to state-of-the-art performance for human detection on the PASCAL datasets.
@article{arxiv.1407.0717,
title = {Deep Poselets for Human Detection},
author = {Lubomir Bourdev and Fei Yang and Rob Fergus},
journal= {arXiv preprint arXiv:1407.0717},
year = {2014}
}