Snap Inc. Image Compression Using GANs
Summary
The USPTO granted Patent US12608593B2 to Snap Inc. on April 21, 2026, covering an image compression system using generative adversarial networks (GANs). The system generates a first GAN, identifies a threshold, generates a second pruned GAN, trains it via similarity-based knowledge distillation, and stores the trained model. The patent contains 20 claims and names five inventors: Jian Ren, Oliver Woodford, Sergey Tulyakov, Jiazhuo Wang, and Qing Jin.
“Systems and methods herein describe an image compression system.”
About this source
GovPing monitors USPTO Patent Grants - AI & Computing (G06N) for new telecom & technology regulatory changes. Every update since tracking began is archived, classified, and available as free RSS or email alerts — 32 changes logged to date.
What changed
USPTO granted Patent US12608593B2 to Snap Inc. on April 21, 2026, covering an image compression system using two generative adversarial networks (GANs). The system generates a first GAN, identifies a compression threshold, prunes channels from the first GAN to produce a second, more efficient GAN, trains the second GAN using similarity-based knowledge distillation, and stores the trained model.
For technology companies and AI developers, this patent represents Snap Inc.'s intellectual property protection in GAN compression techniques. Companies developing similar image-to-image model compression methods should review this patent for potential licensing implications or to assess whether their own approaches may overlap with the 20 claims granted.
Archived snapshot
Apr 23, 2026GovPing captured this document from the original source. If the source has since changed or been removed, this is the text as it existed at that time.
Compressing image-to-image models
Grant US12608593B2 Kind: B2 Apr 21, 2026
Assignee
Snap Inc.
Inventors
Jian Ren, Oliver Woodford, Sergey Tulyakov, Jiazhuo Wang, Qing Jin
Abstract
Systems and methods herein describe an image compression system. The image compression system generates a first generative adversarial network (GAN), identifies a threshold, based on the threshold, generates a second GAN by pruning channels of the first GAN, trains the second GAN using similarity-based knowledge distillation from the first GAN, and stores the trained second GAN.
CPC Classifications
G06N 3/045 G06N 3/088
Filing Date
2021-12-21
Application No.
17558327
Claims
20
Parties
Related changes
Get daily alerts for USPTO Patent Grants - AI & Computing (G06N)
Daily digest delivered to your inbox.
Free. Unsubscribe anytime.
Source
About this page
Every important government, regulator, and court update from around the world. One place. Real-time. Free. Our mission
Source document text, dates, docket IDs, and authority are extracted directly from USPTO.
The summary, classification, recommended actions, deadlines, and penalty information are AI-generated from the original text and may contain errors. Always verify against the source document.
Classification
Who this affects
Taxonomy
Browse Categories
Get alerts for this source
We'll email you when USPTO Patent Grants - AI & Computing (G06N) publishes new changes.
Subscribed!
Optional. Filters your digest to exactly the updates that matter to you.