Modeling the human structure is central for human parsing that extracts pixel-wise semantic
information from images. We start with analyzing three types of inference processes over the …

Human parsing is for pixel-wise human semantic understanding. As human bodies are
underlying hierarchically structured, how to model human structures is the central theme in …

As well known, fully convolutional network (FCN) becomes the state of the art for semantic
segmentation in deep learning. Currently, new hardware designs for deep learning have …

Low cost and easy to use monocular vision systems are able to capture large scale, dense
data in orchards, to facilitate precision agriculture applications. Accurate image parsing is …

We propose an efficient online video filtering method, called adaptive exponential filtering
(AES) to refine pixel prediction maps. Assuming each pixel is associated with a …

Parsing road scene images taken from vehicle mounted cameras provides important
information for high level tasks in automated on-road vehicles. In this paper we adopt the …

When parsing images with regular spatial layout, the location of a pixel (x, y) can provide
important prior for its semantic label. This paper proposes a technique to leverage both …

We propose an algorithm that provides a pixel-wise classification of building facades.
Building facades provide a rich environment for testing semantic segmentation techniques …

The work presented in this dissertation is a step towards effectively incorporating contextual
knowledge in the task of semantic segmentation. To date, the use of context has been …

This thesis studies the problem of semantic segmentation in both images and videos. It
targets at addressing two challenges of semantic segmentation: robustness and …