# Large-scale interactive object segmentation with human annotators

## TL;DR

This paper advances interactive object segmentation by exploring model design, creating a large annotated dataset with human annotators, and developing a method to estimate mask quality automatically.

## Contribution

It systematically studies deep interactive segmentation models, produces the largest instance segmentation dataset, and introduces a technique for automatic mask quality estimation.

## Key findings

- Re-annotating COCO masks is 3x faster than traditional methods.
- Produced 2.5 million instance masks for OpenImages.
- Developed a method to estimate mask quality from annotation signals.

## Abstract

Manually annotating object segmentation masks is very time consuming. Interactive object segmentation methods offer a more efficient alternative where a human annotator and a machine segmentation model collaborate. In this paper we make several contributions to interactive segmentation: (1) we systematically explore in simulation the design space of deep interactive segmentation models and report new insights and caveats; (2) we execute a large-scale annotation campaign with real human annotators, producing masks for 2.5M instances on the OpenImages dataset. We plan to release this data publicly, forming the largest existing dataset for instance segmentation. Moreover, by re-annotating part of the COCO dataset, we show that we can produce instance masks 3 times faster than traditional polygon drawing tools while also providing better quality. (3) We present a technique for automatically estimating the quality of the produced masks which exploits indirect signals from the annotation process.