Cascaded Sparse Feature Propagation Network for Interactive Segmentation

TL;DR

This paper introduces a cascade sparse feature propagation network that efficiently propagates user annotations for high-resolution, detailed interactive segmentation, outperforming existing methods in accuracy and computational efficiency.

Contribution

We propose a novel cascade sparse feature propagation network that improves interactive segmentation by efficiently propagating user annotations with fine-grained detail.

Findings

Outperforms existing methods on various benchmarks.

Enables efficient high-resolution feature propagation.

Achieves more detailed segmentation results.

Abstract

We aim to tackle the problem of point-based interactive segmentation, in which the key challenge is to propagate the user-provided annotations to unlabeled regions efficiently. Existing methods tackle this challenge by utilizing computationally expensive fully connected graphs or transformer architectures that sacrifice important fine-grained information required for accurate segmentation. To overcome these limitations, we propose a cascade sparse feature propagation network that learns a click-augmented feature representation for propagating user-provided information to unlabeled regions. The sparse design of our network enables efficient information propagation on high-resolution features, resulting in more detailed object segmentation. We validate the effectiveness of our method through comprehensive experiments on various benchmarks, and the results demonstrate the superior performance of our approach. Code is available at \href{https://github.com/kleinzcy/CSFPN}{https://github.com/kleinzcy/CSFPN}.