Iterative Learning for Instance Segmentation

TL;DR

This paper introduces an iterative learning method for instance segmentation that reduces annotation effort by requiring minimal human input and works effectively on datasets with multiple similar objects.

Contribution

It presents a novel iterative learning and annotation approach that operates with minimal human intervention and small initial labeled datasets.

Findings

Effective on datasets with multiple similar objects

Requires minimal human annotation

Validates across different visual inspection applications

Abstract

Instance segmentation is a computer vision task where separate objects in an image are detected and segmented. State-of-the-art deep neural network models require large amounts of labeled data in order to perform well in this task. Making these annotations is time-consuming. We propose for the first time, an iterative learning and annotation method that is able to detect, segment and annotate instances in datasets composed of multiple similar objects. The approach requires minimal human intervention and needs only a bootstrapping set containing very few annotations. Experiments on two different datasets show the validity of the approach in different applications related to visual inspection.