Context-aware Feature Generation for Zero-shot Semantic Segmentation

TL;DR

This paper introduces CaGNet, a novel context-aware feature generation method for zero-shot semantic segmentation that leverages contextual information to improve the segmentation of unseen objects without requiring annotations.

Contribution

It proposes a context-aware feature generation approach that enhances zero-shot segmentation by capturing pixel-wise contextual information to generate diverse features.

Findings

Achieves state-of-the-art results on three benchmark datasets.

Effectively leverages semantic word embeddings for unseen object segmentation.

Demonstrates the importance of contextual information in feature generation.

Abstract

Existing semantic segmentation models heavily rely on dense pixel-wise annotations. To reduce the annotation pressure, we focus on a challenging task named zero-shot semantic segmentation, which aims to segment unseen objects with zero annotations. This task can be accomplished by transferring knowledge across categories via semantic word embeddings. In this paper, we propose a novel context-aware feature generation method for zero-shot segmentation named CaGNet. In particular, with the observation that a pixel-wise feature highly depends on its contextual information, we insert a contextual module in a segmentation network to capture the pixel-wise contextual information, which guides the process of generating more diverse and context-aware features from semantic word embeddings. Our method achieves state-of-the-art results on three benchmark datasets for zero-shot segmentation. Codes are available at: https://github.com/bcmi/CaGNet-Zero-Shot-Semantic-Segmentation.