MMLab@NTU was formed on the 1 August 2018. It is now a group with four faculty members and more than 25 members including research fellows, research assistants, and PhD students. Members in MMLab@NTU conduct research primarily in low-level vision, image and video understanding, creative content creation, 3D scene understanding and reconstruction.
Our team is the first to introduce the use of deep neural networks to directly predict super-resolved images. Our journal paper on image super-resolution was selected as the `Most Popular Article' by IEEE Transactions on Pattern Analysis and Machine Intelligence in 2016. It remains as one of the top 10 articles to date. Popular image and video super-resolution methods developed by our team include SRCNN, ESRGAN, EDVR, GLEAN and BasicVSR.
Content Editing and Generation
We research new methods for generating high-resolution, realistic and novel contents in images and videos. We are also interested in investigating fundamental concepts in generative models. Some of our works include scene deocclusion, video inpainting, image generation, and image manipulation.
Image and Video Understanding
We explore effective and efficient methods to detect, segment and recognize objects in complex scenes. We were the champion in COCO 2019 Object Detection Challenge 2019, and Open Images Challenge 2019.
3D Scene Understanding
Our team has been working on various tasks related to 3D reconstruction and perception, e.g, 3D shape generation and 3D human recovery. Check out our recent work on Variational Relational Point Completion Network, Unsupervised 3D Shape Completion through GAN Inversion and LiDAR-based Panoptic Segmentation.
We investigate new deep learning methods that are more efficient, robust, accurate, scalable, transferable, and explainable. We have been working on various problems like domain generalization, knowledge distillation, long-tailed recognition, and self-supervised learning. We participated in Facebook AI Self-Supervision Challenge 2019, and we won champions of all four tracks. Check out OpenSelfSup, our open-source project for self-supervised learning.
- Three Champions in NTIRE 2021 Challenge on Video Restoration and Enhancement, 2021
Kelvin Cheuk Kit Chan, Shangchen Zhou, Xiangyu Xu
Team members: K. C. K. Chan, S. Zhou, X. Xu, C. C. Loy
- Best neural planning metric (PKL), second runner-up, nuScenes Detection Challenge of 5th AI Driving Olympics, NeurIPS 2020
Team members: W. Zhang, T. Wang, K. Chen, D. Lin, C. C. Loy
- Top 10% Outstanding Reviewer, NeurIPS 2020
- Champion, COCO 2019 Object Detection Challenge (Without External Data), 2019
Jiaqi Wang, Wenwei Zhang, Kai Chen
Team members: J. Wang, W. Zhang, Y. Cao, K. Chen, J. Pang, T. Gong, J. Shi, C. C. Loy, D. Lin
- Champions of all four tracks, Facebook AI Self-Supervision Challenge, 2019
Xiaohang Zhan, Jiahao Xie
Team members: X. Zhan, J. Xie, Z. Liu, Y. S. Ong, C. C. Loy
- Champion, Open Images Challenge (Object Detection), 2019
Team members: Y. Liu, G. Song, Y. Zang, Y. Gao, J. Yan, C. C. Loy, X. Wang
- Outstanding Reviewer, ICCV 2019
- Outstanding Reviewer Honorable Mention, BMVC 2019
- Outstanding Reviewers, CVPR 2019
Xintao Wang, Ke Yu
- Champions of all four tracks, NTIRE 2019 Challenge on Video Restoration and Enhancement, 2019
Xintao Wang, Kelvin Cheuk Kit Chan, Ke Yu
Team members: X. Wang, K. C. K. Chan, K. Yu, C. Dong, C. C. Loy
- Champion, PIRM Challenge on Perceptual Super-Resolution (Third Region), 2018
Team members: X. Wang, S. Wu, J. Gu, K. Yu, Y. Liu, C. Dong, Y. Qiao, C. C. Loy
- First Runner-up, DAVIS Challenge on Video Object Segmentation, 2018
Team members: X. Li and C. C. Loy