SEAL uses semantic embeddings and locality-sensitive hashing to create distortion-free, database-free watermarks for generative images that are conditioned on content for improved forgery resistance.
This attack aims to adversarially perturb non-watermarked images such that they appear watermarked by mimicking the latent representation of an originally watermarked image
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SEAL: Semantic Aware Image Watermarking
SEAL uses semantic embeddings and locality-sensitive hashing to create distortion-free, database-free watermarks for generative images that are conditioned on content for improved forgery resistance.