Adobe showcased a series of experimental AI tools at its Max conference, referred to as “sneaks.” These tools aim to enhance the editing of photos, videos, and audio in more intuitive ways. Noteworthy among these is Project Frame Forward, which allows video editors to add or remove elements from footage without the need for traditional masking techniques. The tool can identify and select objects in the first frame of a video, enabling users to remove them and replace them with a similar background. This process is designed to apply seamlessly across the entire video.
Another feature demonstrated was the ability to insert objects into video frames. Users can draw placements and describe the desired new items using AI prompts, which are then integrated contextually throughout the video. An example showcased involved a generated puddle that reflected the movements of an existing cat in the footage.
Project Light Touch utilizes generative AI to manipulate light sources in photographs, allowing users to change the direction and quality of lighting. This tool can create the illusion of additional light sources, control shadows, and even transition environments from day to night, with adjustable color options available for the effects.
Project Clean Take is a new audio editing tool that allows manipulation of speech without necessitating re-recordings. Users can modify voice delivery or emotion, replace specific words, and isolate background noises for targeted adjustments, enhancing voice clarity.
These AI tools, while promising, are currently experimental and not publicly available. They may or may not become standard features in Adobe’s Creative Cloud software or Firefly applications. Previous features, such as Photoshop’s Distraction Removal and Harmonize tools, originated from similar experimental projects, indicating a potential for future availability of these innovations.
Source: https://www.theverge.com/news/811602/adobe-max-2025-sneaks-projects