Wisdom of (Binned) Crowds: A Bayesian Stratification Paradigm for Crowd Counting

The datasets employed to train deep networks for crowd counting are typically heavy-tailed in count distribution and exhibit discontinuities across the count range. As a result, the de-facto statistical measures (MSE, MAE) exhibit large variance and tend to be unreliable indicators of performance across the count range. To address these concerns in a holistic manner, we revise processes at various stages of a standard processing pipeline for crowd counting. To enable principled and balanced minibatch sampling, we propose a novel smoothed Bayesian sample stratification approach. We propose a novel cost function which can be readily incorporated into existing crowd counting deep networks to encourage strataaware optimization. We analyze the performance of representative crowd counting approaches across standard datasets at per strata level and in aggregate. Altogether, our contributions represent a nuanced, statistically balanced and fine-grained characterization of performance for crowd counting approaches.

This zip consists of the pretrained models which can be used for testing. For detailed procedure visit our github repo https://github.com/atmacvit/bincrowd .