CS3H introduces a collision-resistant single-pass framework using normalized Hamming distance loss and attention to improve separation in unsupervised fine-grained image hashing.
(1) A smallerP collision indicates stronger collision resistance, with the ideal caseP collision = 2 −l corresponding to a random l-bit hash function
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Collision-Resistant Single-Pass Method for Unsupervised Fine-Grained Image Hashing
CS3H introduces a collision-resistant single-pass framework using normalized Hamming distance loss and attention to improve separation in unsupervised fine-grained image hashing.