Keshigeyan Chandrasegaran

Hey! I'm a second-year Computer Science PhD student at Stanford University, working on computer vision and machine learning. I’m advised by Prof. Fei-Fei Li and Prof. Juan Carlos Niebles. I am partly supported by a Stanford School of Engineering Fellowship.

Prior to this, I was a Senior Researcher at Temasek Laboratories (SUTD), where I was fortunate to be advised by Prof. Ngai-Man Cheung. Before that, I obtained my Bachelor's Degree in Computer Science, summa cum laude (highest honors) from Singapore University of Technology and Design (SUTD) in 2019.

My research interests are generative modelling, model compression and knowledge transfer. I have also worked with visual geo-localization systems and medical imaging.

Email / GitHub / Google Scholar / Twitter

News

09/24 : One paper accepted to NeurIPS'24!
03/24 : One paper accepted to CVPR'24!
11/23 : Top Reviewer, NeurIPS'23.
09/23 : One paper accepted to NeurIPS'23!
09/23 : After an incredible journey at SUTD, I've left to pursue my PhD at Stanford University!
07/23 : Preprint of our Survey on Generative Modeling with Limited Data, Few Shots, and Zero Shot is now available!
03/23 : One paper accepted to CVPR'23!
02/23 : Invited Talk at NVIDIA on Discovering Transferable Forensic Features [Slides]
10/22 : Top Reviewer, NeurIPS'22
09/22 : One paper accepted to NeurIPS'22.
07/22 : One paper accepted to ECCV'22 as Oral!
07/22 : Invited Talk at NVIDIA on Revisiting Label Smoothing & Knowledge Distillation Compatibility [Slides]
05/22 : One paper accepted to ICML'22!
05/21 : Invited Talk at NVIDIA on Fourier Spectrum Discrepancies for CNN-generated Images Detection [Slides]
03/21 : One paper accepted to CVPR'21 as Oral!

Selected Publications

(*) = Equal contribution

	HourVideo: 1-Hour Video-Language Understanding Keshigeyan Chandrasegaran, Agrim Gupta, Lea M. Hadzic, Taran Kota, Jimming He, Cristobal Eyzaguirre, Zane Durante, Manling Li, Jiajun Wu, Li Fei-Fei NeurIPS Datasets and Benchmarks, 2024. arXiv / project /
	Model Inversion Robustness: Can Transfer Learning Help? Sy-Tuyen Ho, Koh Jun Hao, Keshigeyan Chandrasegaran, Ngoc-Bao Nguyen, Ngai-Man Cheung CVPR, 2024. arXiv / project / code /
	Label-Only Model Inversion Attacks via Knowledge Transfer Ngoc-Bao Nguyen (), Keshigeyan Chandrasegaran (), Milad Abdollahzadeh, Ngai-Man Cheung NeurIPS, 2023. arXiv / project / code / poster / slides
	A Survey on Generative Modeling with Limited Data, Few Shots, and Zero Shot Milad Abdollahzadeh, Touba Malekzadeh (), Christopher T. H. Teo (), Keshigeyan Chandrasegaran (), Guimeng Liu, Ngai-Man Cheung Preprint*, 2023. arXiv / project / code / interactive sankey
	Re-thinking Model Inversion Attacks Against Deep Neural Networks Ngoc-Bao Nguyen (), Keshigeyan Chandrasegaran (), Milad Abdollahzadeh, Ngai-Man Cheung CVPR, 2023. arXiv / project / code / poster / slides
	Few-shot Image Generation via Adaptation-Aware Kernel Modulation Yunqing Zhao (), Keshigeyan Chandrasegaran (), Milad Abdollahzadeh (), Ngai-Man Cheung NeurIPS*, 2022. arXiv / project / code / poster / slides
	Discovering Transferable Forensic Features for CNN-generated Images Detection Keshigeyan Chandrasegaran, Ngoc-Trung Tran, Alexander Binder, Ngai-Man Cheung ECCV, 2022. Oral Presentation arXiv / project / code / poster / slides
	Revisiting Label Smoothing and Knowledge Distillation Compatibility: What was Missing? Keshigeyan Chandrasegaran, Ngoc-Trung Tran (), Yunqing Zhao (), Ngai-Man Cheung ICML, 2022. arXiv / project / code / poster / slides / talk
	A Closer Look at Fourier Spectrum Discrepancies for CNN-generated Images Detection Keshigeyan Chandrasegaran, Ngoc-Trung Tran, Ngai-Man Cheung CVPR, 2021. Oral Presentation arXiv / project / code / slides