ProgRoCC trains a CLIP model with approximate count labels, predicts counts digit by digit (hundreds, tens, units), and beats prior weakly and semi-supervised crowd counters on SHA, QNRF, and JHU++.
Redesigning multi-scale neural network for crowd counting
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
-
ProgRoCC: A Progressive Approach to Rough Crowd Counting
ProgRoCC trains a CLIP model with approximate count labels, predicts counts digit by digit (hundreds, tens, units), and beats prior weakly and semi-supervised crowd counters on SHA, QNRF, and JHU++.