Conference Paper: Semantic Segmentation Learning for Autonomous UAVs using Simulators and Real Data

sos

Abstract: Deep learning requires large amounts of data for training models. For the task of semantic segmentation, manual annotation is time-consuming and difficult. With the recent advances in game engines, simulators have become more popular as they can instantly generate ground truth data for multiple sensors. In this paper, we make a thorough survey of the most recent and popular simulators and synthetic datasets, exploring solutions for semantic segmentation on images taken from drones. We also propose an extension of the CARLA simulator by introducing an aerial camera. We evaluate a deep learning model trained on both synthetic and real data and present a new dataset which comprises both.

Paper download link: Semantic Segmentation Learning for Autonomous UAVs using Simulators and Real Data

ResearchGate link: paper

Optical neural network

InĀ Optica, The Optical Society’s journal for high-impact research, researchers from The Hong Kong University of Science and Technology, Hong Kong detail their two-layer all-optical neural network and successfully apply it to a complex classification task. In conventional hybrid optical neural networks, optical components are typically used for linear operations while nonlinear activation functions — the functions that simulate the way neurons in the human brain respond — are usually implemented electronically because nonlinear optics typically require high-power lasers that are difficult to implement in an optical neural network.

optical-neural-net

Fully functioned 2-layer all optical neural network (AONN)

To overcome this challenge, the researchers used cold atoms with electromagnetically induced transparency to perform nonlinear functions. “This light-induced effect can be achieved with very weak laser power,” said Shengwang Du, a member of the research team. “Because this effect is based on nonlinear quantum interference, it might be possible to extend our system into a quantum neural network that could solve problems intractable by classical methods.”

To confirm the capability and feasibility of the new approach, the researchers constructed a two-layer fully-connected all optical neural network with 16 inputs and two outputs. The researchers used their all-optical network to classify the order and disorder phases of the Ising model, a statistical model of magnetism. The results showed that the all-optical neural network was as accurate as a well-trained computer-based neural network.

Source