Name: Wouter Van Gansbeke
Type: User
Company: KU Leuven
Bio: PhD researcher at KU Leuven.
Especially interested in computer vision, machine learning and deep learning. Working on self-supervised and multi-task learning.
Twitter: WGansbeke
Location: Belgium, Leuven
Blog: https://wvangansbeke.github.io/
Wouter Van Gansbeke's Projects
Build a cross-talk canceler and a speech recognizer
A curated list of awesome self-supervised methods
VIM and TMUX configs
Convert C files into Verilog
End-to-end Lane Detection for Self-Driving Cars (ICCV 2019 Workshop)
Discovering Object Masks for Unsupervised Semantic Segmentation [2022]
Revisiting Contrastive Methods for Unsupervised Learning of Visual Representations. [NeurIPS 2021]
📜 Self-Supervised Learning from Images: Up-to-date reading list.
RSA alogrithm with hardware/software co-design
Predict dense depth maps from sparse and noisy LiDAR frames guided by RGB images. (Ranked 1st place on KITTI) [2019]
Subband filtering with ADPCM
SCAN: Learning to Classify Images without Labels, incl. SimCLR. [ECCV 2020]
Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals. [ICCV 2021]