Zhiwu Huang

I am a Lecturer (Assistant Professor) affiliated with the Vision, Learning, and Control (VLC) group in the School of Electronics and Computer Science (ECS) at the University of Southampton.

I specialize in computer vision and machine learning for artificial general intelligence. I am committed to teaching AI/machines to comprehend the world through multimodal data, enabling them to achieve animal-level, or human-level, or superhuman-level general intelligence. My current focus is on making machine learning models more capable and controllable to understand the physical world, through scaling and aligning compute, data, models, and tasks for deep learning. I am interested in exploring machine learning methodologies, including Riemannian computing, generative modeling, continual learning, and affective computing.

Tutorials:

(2019) Video Quality Mapping Tutorial - From Image Restoration to Enhancement and Beyond (FIRE), ICCV 2019, Seoul, Korea: slides (co-presented with Danda Pani Paudel)

Presentations:

Continual Learning in Computer Vision: A Path Towards Artificial General Intelligence, Seminar Talk, Vision, Learning and Vision (VLC) group, University of Southampton, Feb 9th, 2023

NTIRE 2020 Challenge on Video Quality Mapping: Methods and Results , Oral Presentation, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) workshop, June 15th, 2020

Sliced Wasserstein Generative Models for Image & Video Generation and Enhancement , Invited Talk, Automatic Control Lab, ETH Zürich, June 13th, 2019

Deep Learning on Lie Groups for Skeleton-based Action Recognition , Spotlight Presentation, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 22nd, 2017

Deep Manifold Learning for Human-focussed Video Classification , Postdoc Talk, Computer Vision Lab, ETH Zürich, Nov. 30th, 2016

Log-Euclidean Metric Learning on Symmetric Positive Definite Manifold, Oral Presentation, International Conference on Machine Learning (ICML), July 10th, 2015

Learning Euclidean-to-Riemannian Metric for Point-to-Set Classification, Invited Talk, Vision And Learning Seminar (VALSE) Webinar, Dec. 11th, 2014

Learning Euclidean-to-Riemannian Metric for Point-to-Set Classification, Oral Presentation, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 25th, 2014