Abstract

Generative models are extremely powerful tools to synthesize new images as well as perform various image manipulation tasks. In this project, we study the recent literature on generative modelling and apply them to a variety of image manipulation tasks useful for computer vision applications.

Publications

DGPose: Deep Generative Models for Human Body Analysis.
Rodrigo de Bem, Arnab Ghosh, Thalaiyasingam Ajanthan, Ondrej Miksik, Adnane Boukhayma, N. Siddharth, Philip H.S. Torr.
International Journal of Computer Vision (IJCV), April 2020.
[pdf] [arxiv] [bib]

A Conditional Deep Generative Model of People in Natural Images.
Rodrigo de Bem, Arnab Ghosh, Adnane Boukhayma, Thalaiyasingam Ajanthan, N. Siddharth, Philip H.S. Torr.
Winter Conference on Applications of Computer Vision (WACV), January 2019.
[pdf] [bib]

A Semi-supervised Deep Generative Model for Human Body Analysis.
Rodrigo de Bem, Arnab Ghosh, Thalaiyasingam Ajanthan, Ondrej Miksik, N. Siddharth, Philip H.S. Torr.
ECCV Workshop on Human Behaviour Understanding, September 2018. (oral)
[pdf] [bib]