associate professor at University of Malaya

Google Scholar
Follow @cs-chan

My research interests include computer vision and machine learning with focus on scene understanding. I am also interested in the interplay between language and vision: generating sentential descriptions about complex scenes. I am/was the founding Chair for the IEEE Computational Intelligence Society (CIS) Malaysia chapter, the organising chair for ACPR in 2015, and general chair for MMSP in 2019 & VCIP in 2013. I am a Senior Member of IEEE, a Chartered Engineer and a Member of IET.

        06/2020: One(1) paper to appear in ICPR-2020.
        04/2020: One(1) paper to appear in IJCAI-2020.
        03/2020: One(1) paper (oral) to appear in ICME-2020.
        02/2020: One(1) paper to appear in CVPR-2020.

I am always interested to hear from prospective research students. Scholarships are available from time to time, contact me to enquire.

Latest Works


Rethinking Privacy Preserving Deep Learning: How to Evaluate and Thwart Privacy Attacks

This paper investigates capabilities of Privacy-Preserving Deep Learning (PPDL) mechanisms against various forms of privacy attacks, and a novel Secret Polarization Network (SPN) is proposed to thwart privacy attacks.

L. Fan, K.W. Ng, C. Ju, T. Zhang, C. Liu, C.S. Chan and Q. Yang.
Technical Report


Deep Polarized Network for Supervised Learning of Accurate Binary Hashing Codes

We proposed a novel learning to hash deep neural network, which uses a differentiable hinge-like loss.

L. Fan, K.W. Ng, C. Ju, T. Zhang and C.S. Chan
IJCAI 2020 (acceptance rate: 592/4717 ~ 12.6%)


Style-Conditioned Music Generation Star

We proposed a new formulation to the VAE that allows users to condition on the style of the generated music.

Y-Q Lim, C. S. Chan and F. Y. Loo
ICME 2020 (oral, acceptance rate: 103/834 ~ 12.35%)
[pdf] [code] [sample music]


On the General Value of Evidence, and Bilingual Scene-Text Visual Question Answering

We proposed a new bilingual scene text + evidence VQA dataset named STE-VQA that is annotated with both English and Chinese QA pairs. Within this, an evidence-based measure of an algorithm’s capacity to reason is also proposed that requires the VQA model to provide a bounding box of the predicted answer.

X. Wang, Y. Liu, C. Shen, C.C. Ng, C. Luo, L. Jin, C. S. Chan, A. van den Hengel and L. Wang
CVPR 2020 (acceptance rate: 1470/6656 ~ 22.09%)
[pdf] [dataset]


Total-Text: Toward Orientation Robustness in Scene Text Detection Star

To facilitate a new text detection research, we introduce the Total-Text dataset that consists of 1555 images with more than 3 different text orientations, one of a kind. Extension of ICDAR 2017 (Oral, acceptance rate: 52/409 ~ 12.7%).

C.K. Ch'ng, C.S. Chan and C-L. Liu
International Journal on Document Analysis and Recognition 2020
[pdf] [arXiv (ICDAR 2017 extended ver.)] [slide] [dataset]


Rethinking Deep Neural Network Ownership Verification Star

We propose novel passport-based ownership verification schemes where the deep learning model performance of an original task will be significantly deteriorated due to forged passports.

L. Fan, K.W. Ng and C.S. Chan
NeurIPS 2019 (acceptance rate: 1428/6743 ~ 21.18%)
[pdf] [poster] [code]