RetinaNet Object Detector based on Analog-to-Spiking Neural Network Conversion

TL;DR

This paper presents a novel method to convert complex deep learning object detectors like RetinaNet into spiking neural networks, enabling efficient neuromorphic implementations with minimal performance loss.

Contribution

It introduces a flexible conversion framework capable of handling deep and complex models beyond shallow networks and simple classification tasks.

Findings

RetinaNet can be effectively converted into spiking neural networks

Conversion results in limited performance degradation

The method supports complex, high-performance object detection models

Abstract

The paper proposes a method to convert a deep learning object detector into an equivalent spiking neural network. The aim is to provide a conversion framework that is not constrained to shallow network structures and classification problems as in state-of-the-art conversion libraries. The results show that models of higher complexity, such as the RetinaNet object detector, can be converted with limited loss in performance.