Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Labeling Emotion

Labeling Emotion

Emotional labeling is a cognitive process that involves identifying and naming one's emotions, such as anger, fear, happiness, or sadness. It allows individuals to recognize and express their internal emotional states, a critical aspect of emotional regulation and communication. Labeling emotions requires more than mere recognition; it also involves drawing upon memory and contextual cues to understand the current situation and apply a corresponding emotional label. For instance, feeling...

DC Generator

DC Generator

An alternator converts mechanical energy into electrical energy that varies sinusoidally, resulting in AC current. Meanwhile, a DC generator converts mechanical energy into electrical energy, which are DC pulses with the same polarity. The construction of a DC generator is similar to that of an alternator, except that the pair of slip rings is replaced by a single split ring, also called a commutator. The commutator functions like a periodic rotary switch; it changes the contacts with the...

Modeling and Similitude

Modeling and Similitude

Scaled modeling is a fundamental technique in engineering, enabling the study of large and complex systems by creating smaller, manageable replicas that recreate critical characteristics of the original. In hydrology and civil infrastructure, for example, scaled models of dams help analyze water flow, turbulence, and pressure. This method allows for accurate predictions of real-world behavior within a controlled environment, significantly reducing the cost and time involved in full-scale...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Large language models are powerful electronic health record encoders.

NPJ digital medicine·2026

Same author

Rethinking healthcare data interoperability in the age of large language models.

Med (New York, N.Y.)·2026

Same author

Fish adapt and dynamically avoid an approaching robotic fish across repeated exposures.

Scientific reports·2026

Same author

A predictive atlas of disease onset from retinal fundus photographs: a modelling study using data from population-based cohorts.

The Lancet. Digital health·2026

Same author

Foundation models in oncology win benchmarks but miss the clinic.

Nature cancer·2025

Same author

Honeybees express foodward flight vectors after a detour.

The Journal of experimental biology·2025

Search research articles

Related Experiment Video

Updated: Nov 19, 2025

Photorealistic Learned Landscapes for Augmented Reality

Photorealistic Learned Landscapes for Augmented Reality

Published on: June 27, 2025

RenderGAN: Generating Realistic Labeled Data.

Leon Sixt¹, Benjamin Wild¹, Tim Landgraf¹

¹Fachbereich Mathematik und Informatik, Freie Universität Berlin, Berlin, Germany.

Frontiers in Robotics and AI

|January 27, 2021

Summary

This study introduces RenderGAN, a novel framework for generating realistic, labeled images using 3D models and Generative Adversarial Networks. This approach significantly enhances deep convolutional neuronal network performance in computer vision tasks, overcoming data annotation challenges.

Keywords:

deep learning generative adversarial networks markers social insects unsupervised learning

More Related Videos

Computer-Generated Animal Model Stimuli

Computer-Generated Animal Model Stimuli

Published on: July 29, 2007

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

Related Experiment Videos

Last Updated: Nov 19, 2025

Photorealistic Learned Landscapes for Augmented Reality

Photorealistic Learned Landscapes for Augmented Reality

Published on: June 27, 2025

Computer-Generated Animal Model Stimuli

Computer-Generated Animal Model Stimuli

Published on: July 29, 2007

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

Area of Science:

Computer Vision
Machine Learning
Artificial Intelligence

Background:

Deep Convolutional Neuronal Networks (DCNNs) excel in computer vision but require extensive labeled data.
Manual data annotation is costly and time-consuming, limiting DCNN applicability.
Existing methods struggle to generate sufficient realistic, labeled training data.

Purpose of the Study:

To develop a novel framework, RenderGAN, for generating large volumes of realistic, labeled images.
To address the data scarcity and annotation cost challenges in training DCNNs.
To improve DCNN performance by leveraging synthetic data generation.

Main Methods:

RenderGAN combines 3D models with Generative Adversarial Networks (GANs).
Learns image augmentations (lighting, background, detail) from unlabeled data.
Generates realistic images with preserved labels derived from 3D models.

Main Results:

RenderGAN successfully generated realistic, labeled images of barcode-like markers on honeybees.
DCNNs trained on RenderGAN-generated data significantly outperformed those trained on baselines.
The framework demonstrated the feasibility of using synthetic data for supervised learning.

Conclusions:

RenderGAN offers an effective solution for generating labeled training data for DCNNs.
The framework overcomes limitations associated with manual data annotation.
This approach holds promise for advancing computer vision applications where labeled data is scarce.