Last updated on November 24, 2020
Last week, computer scientist Russell Kirsch, best known for inventing the pixel, passed away. Kirsch and his team of researchers laid the groundwork for all image processing and image pattern recognition. It is fitting this week to focus on news about an exciting breakthrough in image processing, and steps companies are making to digitally mark images that have been altered.
- AI Magic Makes Century-Old Films Look New [Wired] “You can’t call these restorations of films, because the algorithms aren’t just getting rid of imperfections—they’re actually filling in approximations of the data missing from old, blurry, low-frame-rate films. Basically, the algorithms are making stuff up based on their previous training.”
- Adobe’s plans for an online content attribution standard could have big implications for misinformation [TechCrunch] “It will provide a more robust way for content creators to keep their names attached to the work they make. But even more compelling is the idea that the project could provide a technical solution to image-based misinformation. A way to track the provenance of the pictures and videos we encounter online could create a chain of custody that we lack now.”
- Photoshop’s upcoming tagging system will help identify edited images [Engadget] “Last year, with help from The New York Times and Twitter, Adobe started working on the Content Authenticity Initiative, an attempt to cut down on the number of altered images that circulate online. Adobe said the technology would use metadata tagging and cryptography to help the public properly attribute and verify the authenticity of images, videos and other content. That tech will soon get its first test.”
- Adobe plans to preview its system of flagging ‘photoshopped’ images later this year [Neowin] “Under the initiative, Adobe aims to use a system of tags to trace back a given image to the photographer and the location where the photograph was taken. These tags will have a layer of additional security with the help of cryptographic signatures. Whenever a photo is edited, subsequent tags will be added to create a record containing the complete history and origins of the photograph to verify its integrity.”
From the Ohio Web Library:
- Adobe. “Adobe, The New York Times Company and Twitter Announce Content Authenticity Initiative to Develop Industry Standard for Content Attribution.” Business Wire (English), 2019 Apr. 11AD.
- Canales, Manuel, and Ryan T. Williams. “How Eyes Make Sense of Pixels.” National Geographic, vol. 234, no. 4, Oct. 2018, p. 36.
- “Revealed… How AI Technology Is Colouring Our History.” Daily Mail, July 2020, p. 24.