# Generative One-Shot Learning (GOL): A Semi-Parametric Approach to One-Shot Learning in Autonomous Vision

## TL;DR

This paper introduces GOL, a semi-parametric generative framework that creates synthetic training data from one-shot objects to reduce manual annotation in autonomous driving perception systems.

## Contribution

It presents a novel generative approach for one-shot learning that bypasses manual annotation, specifically tailored for autonomous vision applications.

## Key findings

- GOL effectively generates synthetic data from one-shot objects.
- The approach improves perception system training with minimal data.
- GOL performs well on autonomous driving perception challenges.

## Abstract

Highly Autonomous Driving (HAD) systems rely on deep neural networks for the visual perception of the driving environment. Such networks are trained on large manually annotated databases. In this work, a semi-parametric approach to one-shot learning is proposed, with the aim of bypassing the manual annotation step required for training perceptions systems used in autonomous driving. The proposed generative framework, coined Generative One-Shot Learning (GOL), takes as input single one-shot objects, or generic patterns, and a small set of so-called regularization samples used to drive the generative process. New synthetic data is generated as Pareto optimal solutions from one-shot objects using a set of generalization functions built into a generalization generator. GOL has been evaluated on environment perception challenges encountered in autonomous vision.