Computer vision

Atlas camera optimisation suite

Algolux presented a research paper at CVPR 2020, describing an approach applying a stochastic optimisation method to the problem of optimal hardware-software co-design of cameras and computer vision algorithms

Decoding the dichotomy: Traditional image processing versus deep learning

This article is brought to you by: 

This paper will analyse the benefits and drawbacks of traditional image processing and deep learning. It aims to provide clarity on the subject of deep learning, which can then help data scientists and industries choose the most suitable method depending on the task.

A setup for photometric stereo imaging in which multiple lights are used to illuminate an object from different directions. Credit: Advanced illumination

Guided by the light

Matthew Dale explores the power of computational imaging, all made possible by clever illumination

EPIC meeting on structured light and computer vision

The impact of lidar in automotive has been replicated by the application of structured light approaches in machine and computer vision systems. The ultimate paradigm shift is restore vision to the blind. A discussion between photonics hardware, software and users is needed.

Lessons in training neural nets

Limited data is a common problem when training CNNs in industrial imaging applications. Petra Thanner and Daniel Soukup, from the Austrian Institute of Technology, discuss ways of working with CNNs when data is scarce

Reconstruction of traffic signs with high-resolution colour non-line-of-sight imaging using conventional CMOS cameras sensors

Computer vision for seeing around corners presented at CVPR

Carnegie Mellon University showed a non-line-of-sight imaging technique able to compute millimetre- and micrometre-scale shapes of curved objects

Pages

Subscribe to RSS - Computer vision