University of Birmingham > Talks@bham > Artificial Intelligence and Natural Computation seminars > Unsupervised General Dynamic Scene Reconstruction and Understanding

Unsupervised General Dynamic Scene Reconstruction and Understanding

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Hector Basevi.

Host: Dr Hyung Jin Chang (

Abstract: This talk addresses the problem of reconstructing and understanding complex real-world dynamic scenes without prior knowledge of the scene structure, dynamic objects or background. Previous approaches to 3D reconstruction of dynamic scenes require strong initial priors and are limited to relative simple scenes. This talk will present a set of algorithms that enables creating 3D content from videos, creating 4D models from 3D content, application to virtual reality and also understand general dynamic scenes captured from multiple static/moving cameras. The result and evaluation of the proposed methods will be demonstrated on a variety of complex indoor and outdoor scenes demonstrating significant improvement in segmentation, reconstruction and motion estimation.

Biography: Dr. Armin Mustafa is currently a Royal Academy of Engineering Research Fellow since 2018 in the Centre for Vision, Speech and Signal Processing (CVSSP), University of Surrey working in 4D vision for perceptive machines. She finished PhD in general dynamic scene reconstruction from multi-view videos in 2016 from the University of Surrey, after which she became a Research Fellow at CVSSP . She previously worked at Samsung Research Institute , Bangalore, India for 3 years (2010 – 2013) in Computer Vision. In 2010 she received M.Tech. degree from the Indian Institute of Technology (IIT), Kanpur, India. Her research aims to better understand complex scenes so that machines can efficiently model and interpret the real-world for a range of socially beneficial applications including autonomous systems, augmented reality and healthcare. Her research interests include 3D/4D Computer Vision, Scene Understanding, Segmentation, and Multi-view Performance Capture.


This talk is part of the Artificial Intelligence and Natural Computation seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


Talks@bham, University of Birmingham. Contact Us | Help and Documentation | Privacy and Publicity.
talks@bham is based on from the University of Cambridge.