William Thong

PhD candidate at University of Amsterdam

I am a PhD candidate at the Video & Image Sense lab of the University of Amsterdam, under the supervision of Cees Snoek.

Previously, I received a B.Eng. and an M.Sc. in Biomedical Engineering from Polytechnique Montréal. I completed my Master's thesis under the supervision of Samuel Kadoury (MedICAL lab) and Chris Pal (Mila) on the classification of biomedical images with deep learning. I also received an M2 in Bioimaging (BME-Paris) from Télécom Paris.

My main research interests involve visual search, learning with limited labels, and model biases.

[Email]  [Github]  [Google Scholar]  [LinkedIn]

News

[May 2021] I have been recognized as an outstanding reviewer  at CVPR 2021.

 [Apr 2021] Our paper on diverse visual product search  has been accepted in ACM TOMM.

[Mar 2021] Our paper on object priors for zero-shot action recognition  has been accepted in IJCV.

 [Sep 2020] I have been recognized as an outstanding reviewer  at BMVC 2020.

 [Sep 2020] Our paper on model biases in GZSL  has been published at BMVC 2020.

[Jul 2020] Our paper on open cross-domain visual search  has been accepted in CVIU.

Publications

Diversely-Supervised Visual Product Search
William Thong and Cees G.M. Snoek
ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), 2021
[paper]  [arxiv]  [code

We create a diverse set of labels from instance, attribute and category similarities for visual product search.

Object Priors for Classifying and Localizing Unseen Actions
Pascal Mettes, William Thong, Cees G.M. Snoek
International Journal of Computer Vision (IJCV), 2021
[paper]  [arxiv]  [code

We derive spatial and semantic priors to recognize unseen actions with zero training samples.

Bias-Awareness for Zero-Shot Learning the Seen and Unseen
William Thong and Cees G.M. Snoek
British Machine Vision Conference (BMVC), 2020
[paper]  [arxiv]  [code]  [video

We mitigate the classifier bias towards classes seen during training in generalized zero-shot learning.

Open Cross-Domain Visual Search
William Thong, Pascal Mettes, Cees G.M. Snoek
Computer Vision and Image Understanding (CVIU), 2020
[paper]  [arxiv]  [code

We search for categories from any source domain to any target domain in a common semantic space.

A Layer-Based Sequential Framework for Scene Generation with GANs
Mehmet O. Turkoglu, William Thong, Luuk Spreeuwers, Berkay Kicanaoglu
AAAI Conference on Artificial Intelligence (AAAI), 2019
[paper]  [arxiv]  [poster]  [code]

We compose a scene layer-by-layer, with an explicit control over the generation of all scene elements.

Convolutional Networks for Kidney Segmentation in Contrast-Enhanced CT Scans
William Thong, Samuel Kadoury, Nicolas Piché, Christopher J. Pal
CMBBE: Imaging & Visualization, 2018
[paper] – initially presented at MICCAI-DLMIA 2015

We segment healthy and abnormal kidneys in CT scans with a patch-based ConvNet.

Three-dimensional Morphology Study of Surgical Adolescent Idiopathic Scoliosis Patient from Encoded Geometric Models
William Thong, Stefan Parent, James Wu, Carl-Éric Aubin, Hubert Labelle, Samuel Kadoury
European Spine Journal (ESJ), 2016
[paper]

We cluster the predominant modes of variability of scoliotic spine deformations from all Lenke types with a stacked auto-encoder.

Automatic Labeling of Vertebral Levels using a Robust Template-Based Approach
Eugénie Ullmann, Jean François Pelletier Paquette*, William Thong*, Julien Cohen-Adad
International Journal of Biomedical Imaging (IJBI), 2014
[paper]

We build a template to predict vertebral levels in MRI images.

Workshop papers & Abstracts

Interactive Exploration of Journalistic Video Footage through Multimodal Semantic Matching
S Ibrahimi, S Chen, D Arya, A Camara, Y Chen, T Crijns, M Van der Goes, T Mensink, E Van Miltenburg, D Odijk, W Thong, J Zhao, P Mettes
ACM Multimedia (Demo track), 2019
[paper]

Stacked Auto-Encoders for Classification of 3D Spine Models in Adolescent Idiopathic Scoliosis
William Thong, Hubert Labelle, Jesse Shen, Stefan Parent, Samuel Kadoury
Recent Advances in Computational Methods and Clinical Applications for Spine Imaging, 2015
[paper] – initially presented at MICCAI-CSI 2014

Spinal Cord Toolbox: an Open-Source Framework for Processing Spinal Cord MRI Data
J Cohen-Adad, B De Leener, M Benhamou, D Cadotte, D Fleet, A Cadotte, MG Fehlings, JF Pelletier Paquette, W Thong, M Taso, DL Collins, V Callot, V Fonov
OHBM, 2014
[poster]


Webpage template from this researcher.