Shadows Generation in the Wild
Talk video
Talk presentation
The presence of strong illumination variations in images, shadows, in particular, have been shown to be problematic for a variety of computer vision algorithms such as object recognition and recognition.
We'll briefly talk about shadow detection, removal and focus on shadow generation in the wild for data augmentation.
Taras Lehinevych
Rails Reactor
- Machine Learning Engineer at Rails Reactor
- Lecturer at National University of Kyiv-Mohyla Academy and Rails Reactor Machine Learning Summer School
- PhD student in Computer Science at the IPS NAN of Ukraine and NaUKMA