ENInst: Enhancing Weakly-supervised Low-shot Instance Segmentation

TL;DR

ENInst is a novel method that improves weakly-supervised low-shot instance segmentation by enhancing sub-tasks, achieving high efficiency and competitive performance compared to fully-supervised models.

Contribution

The paper introduces ENInst, a new approach with sub-task enhancement techniques for better weakly-supervised low-shot instance segmentation.

Findings

ENInst is 7.5 times more efficient than existing methods.

ENInst outperforms some fully-supervised few-shot models.

Systematic analysis identified key bottlenecks in the problem.

Abstract

We address a weakly-supervised low-shot instance segmentation, an annotation-efficient training method to deal with novel classes effectively. Since it is an under-explored problem, we first investigate the difficulty of the problem and identify the performance bottleneck by conducting systematic analyses of model components and individual sub-tasks with a simple baseline model. Based on the analyses, we propose ENInst with sub-task enhancement methods: instance-wise mask refinement for enhancing pixel localization quality and novel classifier composition for improving classification accuracy. Our proposed method lifts the overall performance by enhancing the performance of each sub-task. We demonstrate that our ENInst is 7.5 times more efficient in achieving comparable performance to the existing fully-supervised few-shot models and even outperforms them at times.