The Feed30.09.2026community.adobe.com

Firefly removes video backgrounds without frame-by-frame masks

Firefly can strip backgrounds across an entire clip, with longer prompts and wider GPT Image 2 ratios.

Firefly removes video backgrounds without frame-by-frame masks

Adobe's 18 September Firefly update leads with Remove Background for video: automatic subject detection across a whole clip instead of frame-by-frame masking. The same drop widens GPT Image 2 to 10 aspect ratios on web and Boards, adds reference images for Kling 3.0 Omni in Video Generation and Boards, and lengthens prompts for image and video so one idea stays in one credit path.

What this means for creatives is fewer round trips for cutdowns, product float and social variants. It does not retire compositing craft on hard edges and hair. It moves the first 80% of cleanup into the Firefly surface teams already open.

The FeedNext storyGPT-6 Sol and Luna cut the price of everyday agent craft

Stay ahead of creative AI.

New editions when they drop. The best of the Feed in between. One email, no noise.