Neural Networks Easily Fooled by Common Objects Seen from New Angles

New research shows how AI networks often misidentify common objects such as school buses when seen from different perspectives.

Image Forensics: AI Can Detect Facial Manipulations in Photos

Adobe Research is now exploring using new technologies, including AI methods, to identify manipulated imagery and to increase trust and authority in digital media.

PhotoGeode: 3D Faces from 2D Imagery

PhotoGeode, an Adobe Research project spearheaded by TJ Rhodes, reconstructs a 3D model of a person's face using 2D photos.

Augmenting Live Videos with Interactive Graphics

An Adobe Research team is exploring how to enhance real-time presentations with interactive graphics to create a powerful new storytelling environment.

Content-Aware Fill for Video Wins Customers’ Hearts—and Broadcasters’ Awards

Content-Aware Fill for Video, created by Adobe Research, empowers editors to automatically remove unwanted objects in their videos in an intuitive way. It's integrated into Adobe’s After Effects software.

How a Ball Bounces: Teaching a Computer Real-World Physics

Adobe Research and collaborators are teaching computers to learn to make predictions about the physics of everyday real-world interactions.

Helping Vloggers Create Better Videos

Adobe Research's intuitive, text-based interface could allow even novice vloggers to quickly create more engaging videos.

Adobe Research Shares Technology Sneaks for Marketers

At this year’s Adobe Summit, Adobe Research took center stage to share demos of future tech offerings for marketers and data analysts.

Adobe TechSummit 2019 in Photos

The 25th anniversary Adobe TechSummit offered a platform to many Adobe Research scientists and engineers to share their work and vision for the future. Here are just a few highlights of the San Francisco event that welcomed more than 3,000 Adobe staff and inspired them to continue to advance innovation....