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

Light Acquisition

Light Acquisition

In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.

Deconvolution

Deconvolution

Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...

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

Controlled Generation of Aldehydes or Carboxylic Acids Driven by Mn-Doped NiO.

Organic letters·2026

Same author

Interpreting the Observed Behavior of a Class of Autonomous Linear Systems Using Explainable Inverse Reinforcement Learning.

IEEE transactions on neural networks and learning systems·2026

Same author

Generative AI for climate governance and acceptability-constrained policy design.

npj climate action·2026

Same author

Organic Electrochemical Cathodic Reduction of CO, CN, and NN.

Chemical record (New York, N.Y.)·2026

Same author

Improving ADM1 predictions via Bayesian analysis for continuous anaerobic digestion.

Journal of environmental management·2026

Same author

Photoelectrocatalytic Chlorothiocyanation of Unactivated Alkenes via Ligand-to-Metal Charge Transfer.

Organic letters·2025

Search research articles

Related Experiment Video

Updated: Jul 3, 2026

Localizing Protein in 3D Neural Stem Cell Culture: a Hybrid Visualization Methodology

Localizing Protein in 3D Neural Stem Cell Culture: a Hybrid Visualization Methodology

Published on: December 19, 2010

Sim2Real: Generative AI to Enhance Photorealism through Domain Transfer with GAN and Seven-Chanel-360°-Paired-Images

Marc Bresson¹, Yang Xing¹, Weisi Guo¹

¹School of Aerospace, Transport and Manufacturing, Cranfield University, Bedfordshire MK43 0AL, UK.

Sensors (Basel, Switzerland)

|January 11, 2024

Summary

This study addresses data scarcity by generating photorealistic images from 3D scenes using advanced Generative Adversarial Networks (GANs). The method creates controllable, high-quality visuals for AI model training.

Keywords:

360° images GAN dataset domain transfer generative AI perspective augmentation photorealism simulation

More Related Videos

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Published on: February 8, 2014

Laser-induced Forward Transfer of Ag Nanopaste

Laser-induced Forward Transfer of Ag Nanopaste

Published on: March 31, 2016

Related Experiment Videos

Last Updated: Jul 3, 2026

Localizing Protein in 3D Neural Stem Cell Culture: a Hybrid Visualization Methodology

Localizing Protein in 3D Neural Stem Cell Culture: a Hybrid Visualization Methodology

Published on: December 19, 2010

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Published on: February 8, 2014

Laser-induced Forward Transfer of Ag Nanopaste

Laser-induced Forward Transfer of Ag Nanopaste

Published on: March 31, 2016

Area of Science:

Computer Vision
Artificial Intelligence
Image Generation

Background:

Data scarcity is a significant challenge in training AI models for image generation.
Generative Adversarial Networks (GANs) are prevalent for image synthesis, but often limited to RGB-to-RGB tasks.
Existing methods lack fine-grained control over generated image content, such as building shapes and environments.

Purpose of the Study:

To develop a solution for data scarcity in image generation by enabling controlled synthesis of new images.
To adapt state-of-the-art GANs for transforming multi-channel 3D scene data into photorealistic images.
To create a method that allows users to control building shapes and environmental elements in generated images.

Main Methods:

Utilized a state-of-the-art Generative Adversarial Network (GAN) for domain transfer.
Developed a custom dataset pairing simulated 3D scene data (seven-channel images including depth, segmentation map, surface normal) with real 360° street views from Paris.
Designed a generator network to preserve semantic information across layers for enhanced realism.

Main Results:

Successfully transformed multi-channel simulated 3D scene data into photorealistic 360° images of Paris.
Generated samples exhibit high realism and maintain semantic consistency.
The generated images are suitable for training and improving future AI models.

Conclusions:

The proposed GAN-based approach effectively addresses data scarcity by generating controllable, photorealistic images from 3D scene data.
The method demonstrates the potential for creating synthetic data that can enhance AI model performance.
Further refinements can improve model performance and expand applicability.