|
I am an Executive Director of SenseTime.
I was a participant on the predictive modelling competition platform Kaggle. My highest Kaggle ranking was 6th.
I obtained the PhD degree from The Chinese University of Hong Kong, and the BEng degree from Tsinghua University, Beijing. I spent a year as a research intern at Microsoft Research Asia. I also spent two years as a quantitative trader, developing high-frequency market-making strategies on equity and future markets with the knowledge of machine learning and market microstructure.
Email:
|
03/2023: Four papers including one highlight are accepted by CVPR 2023.
03/2022: ViM for out-of-distribution detection is accepted by CVPR 2022. It has been included in several courses such as Intro to ML Safety, Recent Advances in Deep Learning, and open source libraries such as OpenOOD, pytorch-ood, Oodeel.
07/2021: Three papers are accepted by ICCV 2021.
03/2021: One paper on scene text detection is accepted by CVPR 2021, and one paper on multi-task learning is accepted by ICLR 2021.
07/2020: Three papers on action detection, text recognition and noisy learning are accepted by ECCV 2020, one paper (oral) on webly learning is accepted by ACM Multimedia 2020.
03/2020: Our paper Scale-equalizing pyramid convolution for object detection is accepted by CVPR 2020.
11/2019: Our paper Object instance mining for weakly supervised object detection is accepted by AAAI 2020. [github]
07/2019: Two papers (1 oral, 1 poster) are accepted by ICCV 2019, two papers (1 oral, 1 poster) are accepted by ACM Multimedia 2019 and one paper is accepted by BMVC 2019.
06/2019: Our team won the first runner-up in WebVision Challenge 2019. [results] [Presentation]
05/2019: Our team won the first runner-up in Scene text recognition (Chinese and Latin) Task of ICDAR2019 Robust Reading Challenge on Arbitrary-Shaped Text (ICDAR 2019-ArT). [results]
03/2019: Our paper Data-driven neuron allocation for scale aggregation networks was accepted by CVPR 2019. [github]