HistoNet: Predicting size histograms of object instances

TL;DR

HistoNet is a novel approach that predicts size histograms of objects in crowded scenes directly, bypassing explicit instance segmentation, which enhances accuracy and reduces complexity for applications like biology and medicine.

Contribution

The paper introduces HistoNet, a new method for directly predicting object size histograms in crowded scenes, along with a new dataset for this task.

Findings

HistoNet outperforms Mask R-CNN in size distribution prediction.

The approach requires fewer parameters than explicit segmentation methods.

Application to biological and medical imaging demonstrates practical utility.

Abstract

We propose to predict histograms of object sizes in crowded scenes directly without any explicit object instance segmentation. What makes this task challenging is the high density of objects (of the same category), which makes instance identification hard. Instead of explicitly segmenting object instances, we show that directly learning histograms of object sizes improves accuracy while using drastically less parameters. This is very useful for application scenarios where explicit, pixel-accurate instance segmentation is not needed, but there lies interest in the overall distribution of instance sizes. Our core applications are in biology, where we estimate the size distribution of soldier fly larvae, and medicine, where we estimate the size distribution of cancer cells as an intermediate step to calculate the tumor cellularity score. Given an image with hundreds of small object instances, we output the total count and the size histogram. We also provide a new data set for this task, the FlyLarvae data set, which consists of 11,000 larvae instances labeled pixel-wise. Our method results in an overall improvement in the count and size distribution prediction as compared to state-of-the-art instance segmentation method Mask R-CNN.