An uncertainty-based sampling method selects a small subset of image pairs for human evaluation and uses a deep model for the rest, preserving accuracy with fewer comparisons.
Comparison of four subjective methods for image quality assessment,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.MM 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Uncertainty-driven Sampling for Efficient Pairwise Comparison Subjective Assessment
An uncertainty-based sampling method selects a small subset of image pairs for human evaluation and uses a deep model for the rest, preserving accuracy with fewer comparisons.