Deep Interactive Object Selection

TL;DR

This paper introduces a deep learning-based interactive object selection method that significantly reduces user interactions by leveraging Euclidean distance maps and deep FCNs, achieving superior results across multiple datasets.

Contribution

A novel deep learning approach that transforms user clicks into distance maps and combines them with RGB images to improve interactive object selection efficiency.

Findings

Reduces user interactions to just a few clicks

Achieves superior accuracy compared to existing methods

Demonstrates strong generalization on unseen object classes

Abstract

Interactive object selection is a very important research problem and has many applications. Previous algorithms require substantial user interactions to estimate the foreground and background distributions. In this paper, we present a novel deep learning based algorithm which has a much better understanding of objectness and thus can reduce user interactions to just a few clicks. Our algorithm transforms user provided positive and negative clicks into two Euclidean distance maps which are then concatenated with the RGB channels of images to compose (image, user interactions) pairs. We generate many of such pairs by combining several random sampling strategies to model user click patterns and use them to fine tune deep Fully Convolutional Networks (FCNs). Finally the output probability maps of our FCN 8s model is integrated with graph cut optimization to refine the boundary segments. Our model is trained on the PASCAL segmentation dataset and evaluated on other datasets with different object classes. Experimental results on both seen and unseen objects clearly demonstrate that our algorithm has a good generalization ability and is superior to all existing interactive object selection approaches.