UG-ICM trains a single learned image codec with global, local, and instance-level CLIP supervision and a preference-conditioned decoder to support both human viewing and unseen machine analytics tasks from one bitstream.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Unified Coding for Both Human Perception and Generalized Machine Analytics with CLIP Supervision
UG-ICM trains a single learned image codec with global, local, and instance-level CLIP supervision and a preference-conditioned decoder to support both human viewing and unseen machine analytics tasks from one bitstream.