Training Deep Networks with Synthetic Data: Bridging the Reality Gap by Domain Randomization

04/18/2018
by   Jonathan Tremblay, et al.
0

We present a system for training deep neural networks for object detection using synthetic images. To handle the variability in real-world data, the system relies upon the technique of domain randomization, in which the parameters of the simulator-such as lighting, pose, object textures, etc.-are randomized in non-realistic ways to force the neural network to learn the essential features of the object of interest. We explore the importance of these parameters, showing that it is possible to produce a network with compelling performance using only non-artistically-generated synthetic data. With additional fine-tuning on real data, the network yields better performance than using real data alone. This result opens up the possibility of using inexpensive synthetic data for training neural networks while avoiding the need to collect large amounts of hand-annotated real-world data or to generate high-fidelity synthetic worlds-both of which remain bottlenecks for many applications. The approach is evaluated on bounding box detection of cars on the KITTI dataset.

READ FULL TEXT

page 3

page 4

page 6

page 7

research
01/31/2022

Reducing the Amount of Real World Data for Object Detector Training with Synthetic Data

A number of studies have investigated the training of neural networks wi...
research
11/11/2021

Training neural networks with synthetic electrocardiograms

We present a method for training neural networks with synthetic electroc...
research
10/23/2018

Structured Domain Randomization: Bridging the Reality Gap by Context-Aware Synthetic Data

We present structured domain randomization (SDR), a variant of domain ra...
research
07/16/2019

How much real data do we actually need: Analyzing object detection performance using synthetic and real data

In recent years, deep learning models have resulted in a huge amount of ...
research
05/23/2019

Watermark retrieval from 3D printed objects via synthetic data training

We present a deep neural network based method for the retrieval of water...
research
07/17/2020

Mixing Real and Synthetic Data to Enhance Neural Network Training – A Review of Current Approaches

Deep neural networks have gained tremendous importance in many computer ...
research
04/25/2019

Improved visible to IR image transformation using synthetic data augmentation with cycle-consistent adversarial networks

Infrared (IR) images are essential to improve the visibility of dark or ...

Please sign up or login with your details

Forgot password? Click here to reset