Articulated Pose Estimation by a Graphical Model with Image Dependent Pairwise Relations
This addresses the problem of accurately estimating human poses in images for computer vision applications, representing an incremental improvement by combining graphical models with deep learning.
The paper tackles articulated human pose estimation from a single image by developing a graphical model with image-dependent pairwise relations, using deep convolutional neural networks to learn part and spatial relationship probabilities, and achieves state-of-the-art performance on LSP and FLIC datasets.
We present a method for estimating articulated human pose from a single static image based on a graphical model with novel pairwise relations that make adaptive use of local image measurements. More precisely, we specify a graphical model for human pose which exploits the fact the local image measurements can be used both to detect parts (or joints) and also to predict the spatial relationships between them (Image Dependent Pairwise Relations). These spatial relationships are represented by a mixture model. We use Deep Convolutional Neural Networks (DCNNs) to learn conditional probabilities for the presence of parts and their spatial relationships within image patches. Hence our model combines the representational flexibility of graphical models with the efficiency and statistical power of DCNNs. Our method significantly outperforms the state of the art methods on the LSP and FLIC datasets and also performs very well on the Buffy dataset without any training.