Search Results for author: Thomas Boulay

Found 3 papers, 0 papers with code

Multi-Task Network Pruning and Embedded Optimization for Real-time Deployment in ADAS

no code implementations19 Jan 2021 Flora Dellinger, Thomas Boulay, Diego Mendoza Barrenechea, Said El-Hachimi, Isabelle Leang, Fabien Bürger

Secondly, a pruning method is applied to compress the CNN, helping to reduce the runtime and memory usage by a factor of 2 without lowering the performances significantly.

Network Pruning object-detection +3

YUVMultiNet: Real-time YUV multi-task CNN for autonomous driving

no code implementations11 Apr 2019 Thomas Boulay, Said El-Hachimi, Mani Kumar Surisetti, Pullarao Maddu, Saranya Kandan

In this paper, we propose a multi-task convolutional neural network (CNN) architecture optimized for a low power automotive grade SoC.

Autonomous Driving Quantization +1

Cannot find the paper you are looking for? You can Submit a new open access paper.