Add to BibSonomyAbstract
We present a wide-angle design simulation using simplified, computationally efficient PSF models which maintain accurate neural network performance predictions. By integrating it into an end-to-end differentiable pipeline, we optimize the distortion for improved depth estimation.
© 2023 The Author(s)
PDF Article | Presentation VideoMore Like This
Wide angle imaging system image simulation for accurate neural network performances prediction.
Julie Buquet, Raphael Larouche, Jocelyn Parent, Jean-François Lalonde, and Simon Thibault
CTh4C.3 Computational Optical Sensing and Imaging (COSI) 2022
Learning-based depth estimation on wide angle images with non linear distortion
Julie Buquet, Jinsong Zhang, Patrice Roulet, Jean-Francois Lalonde, and Simon Thibault
DF3A.1 3D Image Acquisition and Display: Technology, Perception and Applications (3D) 2020
Optimizing Deep Learning Based Retinal Diseases Classification On Optical Coherence Tomography Scans
Aditya Chandra Mandal and Abhijeet Phatak
1263220 European Conference on Biomedical Optics (ECBO) 2023