Edinburgh, United Kingdom

Oisin MacAodha

This inventor holds 4 USPTO granted patents and 1 EPO patent. Top assignees: Niantic, Inc., The University of Edinburgh, Toyota Jidosha Kabushiki Kaisha. Active years: 2023-2024.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Oisin MacAodha. Retrieved from https://idiyas.com/inventor/oisin-macaodha-royhkte2

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile


% Patents Active = 100.0

 

Average Co-Inventor Count = 4.9

ph-index = 1

Forward Citations = 3(Granted Patents)


Location History:

  • London, GB (2023)
  • Edinburgh, GB (2024)

Company Filing History:


Years Active: 2023-2024

Loading Chart...
Loading Chart...
4 patents (USPTO):Explore Patents

Title: Oisin MacAodha: Innovator in Scene Reconstruction and Neural Networks

Introduction

Oisin MacAodha is a prominent inventor based in Edinburgh, GB. He has made significant contributions to the fields of scene reconstruction and neural networks, holding a total of four patents. His innovative work has the potential to enhance various applications in augmented reality and computer vision.

Latest Patents

One of MacAodha's latest patents is titled "High-speed real-time scene reconstruction from input image data." This patent discloses a scene reconstruction model that outputs a heightfield for a series of input images. The model predicts depth maps for each input image and extracts feature maps, ultimately building a 3D model. It utilizes the predicted depth maps and camera poses to determine a raw heightfield for the scene. The model samples features from the feature maps corresponding to positions on the heightfield and aggregates them into an aggregate feature map. It then regresses a refined heightfield based on this aggregate feature map, determining the final heightfield. This innovation allows client devices to generate virtual content augmented on real-world images captured by the device.

Another notable patent is a "Method for training neural network, system for training neural network, and neural network." This method involves training a first neural network to detect the viewpoint of an object visible in an image. It includes providing a dataset of pairs of images under different viewpoints and utilizing two additional neural networks to deliver appearance information and synthetic images of the object. The joint training of these networks enhances the accuracy of viewpoint detection.

Career Highlights

Oisin MacAodha has worked with notable companies such as Niantic, Inc. and Toyota Jidosha Kabushiki Kaisha. His experience in these organizations has contributed to his expertise in developing advanced technologies in the field of computer vision and augmented reality.

Collaborations

Some of his notable coworkers include James Watson and Gabriel J Brostow. Their collaboration has likely fostered an environment of innovation and creativity in their respective projects.

Conclusion

Oisin MacAodha is a distinguished inventor whose work in scene reconstruction and neural networks is paving the way for advancements in augmented reality. His patents reflect a deep understanding of technology and its applications, making him a valuable contributor to the field.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…