SS-CPGAN: Self-Supervised Cut-and-Pasting Generative Adversarial Network for Object Segmentation

Abstract

This paper proposes a novel self-supervised based Cut-and-Paste GAN to perform foreground object segmentation and generate realistic composite images without manual annotations. We accomplish this goal by a simple yet effective self-supervised approach coupled with the U-Net based discriminator. The proposed method extends the ability of the standard discriminators to learn not only the global data representations via classification (real/fake) but also learn semantic and structural information through pseudo labels created using the self-supervised task. The proposed method empowers the generator to create meaningful masks by forcing it to learn informative per-pixel as well as global image feedback from the discriminator. Our experiments demonstrate that our proposed method significantly outperforms the state-of-the-art methods on the standard benchmark datasets.

References in corpus (7)

• Self-Supervised Representation Learning: Introduction, Advances and Challenges
• Unsupervised Learning of Image Segmentation Based on Differentiable Feature Clustering
• A Survey of Natural Language Generation
• Self-supervised GAN: Analysis and Improvement with Multi-class Minimax Game
• Object-aware Contrastive Learning for Debiased Scene Representation
• DeshuffleGAN: A Self-Supervised GAN to Improve Structure Learning
• Object Segmentation Without Labels with Large-Scale Generative Models