Efficient Interactive Annotation of Segmentation Datasets with Polygon-RNN++

TL;DR

Polygon-RNN++ is an advanced interactive annotation model that significantly improves the efficiency and accuracy of polygonal object annotations in images, reducing human effort and generalizing well across datasets.

Contribution

The paper introduces Polygon-RNN++, a novel model with a new CNN encoder, reinforcement learning training, and high-resolution output via GNN, enhancing interactive segmentation annotation.

Findings

Outperforms original Polygon-RNN with 10% higher mean IoU

Requires 50% fewer clicks for annotation

Shows strong cross-domain generalization

Abstract

Manually labeling datasets with object masks is extremely time consuming. In this work, we follow the idea of Polygon-RNN to produce polygonal annotations of objects interactively using humans-in-the-loop. We introduce several important improvements to the model: 1) we design a new CNN encoder architecture, 2) show how to effectively train the model with Reinforcement Learning, and 3) significantly increase the output resolution using a Graph Neural Network, allowing the model to accurately annotate high-resolution objects in images. Extensive evaluation on the Cityscapes dataset shows that our model, which we refer to as Polygon-RNN++, significantly outperforms the original model in both automatic (10% absolute and 16% relative improvement in mean IoU) and interactive modes (requiring 50% fewer clicks by annotators). We further analyze the cross-domain scenario in which our model is trained on one dataset, and used out of the box on datasets from varying domains. The results show that Polygon-RNN++ exhibits powerful generalization capabilities, achieving significant improvements over existing pixel-wise methods. Using simple online fine-tuning we further achieve a high reduction in annotation time for new datasets, moving a step closer towards an interactive annotation tool to be used in practice.