Regularized Densely-connected Pyramid Network for Salient Instance Segmentation

TL;DR

This paper introduces a new end-to-end salient instance segmentation method that enhances feature utilization through regularized dense connections and a multi-level RoIAlign decoder, significantly outperforming existing methods.

Contribution

It proposes regularized dense connections and a multi-level RoIAlign decoder within the Mask R-CNN framework for improved salient instance segmentation.

Findings

Achieves 6.3% higher AP than state-of-the-art methods.

Effectively utilizes multi-level features for better mask prediction.

Outperforms existing methods on popular benchmarks.

Abstract

Much of the recent efforts on salient object detection (SOD) have been devoted to producing accurate saliency maps without being aware of their instance labels. To this end, we propose a new pipeline for end-to-end salient instance segmentation (SIS) that predicts a class-agnostic mask for each detected salient instance. To better use the rich feature hierarchies in deep networks and enhance the side predictions, we propose the regularized dense connections, which attentively promote informative features and suppress non-informative ones from all feature pyramids. A novel multi-level RoIAlign based decoder is introduced to adaptively aggregate multi-level features for better mask predictions. Such strategies can be well-encapsulated into the Mask R-CNN pipeline. Extensive experiments on popular benchmarks demonstrate that our design significantly outperforms existing \sArt competitors by 6.3\% (58.6\% vs. 52.3\%) in terms of the AP metric.The code is available at https://github.com/yuhuan-wu/RDPNet.