A human-vision model chooses the smallest super-resolution network or branch for each image patch, cutting FLOPs by up to 78% while keeping output visually indistinguishable from full-network SR in small user studies.
Mantiuk, Alexandre Chapiro, and Sophie Wuerger
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Human Vision Constrained Super-Resolution
A human-vision model chooses the smallest super-resolution network or branch for each image patch, cutting FLOPs by up to 78% while keeping output visually indistinguishable from full-network SR in small user studies.