Guided Interactive Video Object Segmentation Using Reliability-Based Attention Maps

TL;DR

This paper introduces a guided interactive video object segmentation method that leverages reliability-based attention and intersection-aware propagation to enhance accuracy and efficiency in video segmentation tasks.

Contribution

It presents a novel GIS algorithm with reliability-based attention and intersection-aware propagation modules, improving segmentation accuracy and reducing user interaction time.

Findings

More accurate segmentation results than conventional methods

Faster processing speed in interactive video segmentation

Effective user interaction mechanism for selecting unsatisfactory frames

Abstract

We propose a novel guided interactive segmentation (GIS) algorithm for video objects to improve the segmentation accuracy and reduce the interaction time. First, we design the reliability-based attention module to analyze the reliability of multiple annotated frames. Second, we develop the intersection-aware propagation module to propagate segmentation results to neighboring frames. Third, we introduce the GIS mechanism for a user to select unsatisfactory frames quickly with less effort. Experimental results demonstrate that the proposed algorithm provides more accurate segmentation results at a faster speed than conventional algorithms. Codes are available at https://github.com/yuk6heo/GIS-RAmap.