Towards Partial Supervision for Generic Object Counting in Natural Scenes

TL;DR

This paper introduces two novel frameworks, LC and RLC, for generic object counting in natural scenes under a partially supervised setting, significantly reducing annotation costs while maintaining competitive accuracy.

Contribution

The paper proposes the LC and RLC frameworks that enable object counting with lower supervision levels, including a new dual-branch architecture and a weight modulation layer for class-label prediction.

Findings

Perform on par with state-of-the-art methods with less supervision.

RLC reduces annotation costs for large category sets.

LC framework enables density map generation with minimal supervision.

Abstract

Generic object counting in natural scenes is a challenging computer vision problem. Existing approaches either rely on instance-level supervision or absolute count information to train a generic object counter. We introduce a partially supervised setting that significantly reduces the supervision level required for generic object counting. We propose two novel frameworks, named lower-count (LC) and reduced lower-count (RLC), to enable object counting under this setting. Our frameworks are built on a novel dual-branch architecture that has an image classification and a density branch. Our LC framework reduces the annotation cost due to multiple instances in an image by using only lower-count supervision for all object categories. Our RLC framework further reduces the annotation cost arising from large numbers of object categories in a dataset by only using lower-count supervision for a subset of categories and class-labels for the remaining ones. The RLC framework extends our dual-branch LC framework with a novel weight modulation layer and a category-independent density map prediction. Experiments are performed on COCO, Visual Genome and PASCAL 2007 datasets. Our frameworks perform on par with state-of-the-art approaches using higher levels of supervision. Additionally, we demonstrate the applicability of our LC supervised density map for image-level supervised instance segmentation.