Target-Free Text-guided Image Manipulation

TL;DR

This paper introduces cManiGAN, a novel cyclic GAN framework for target-free, text-guided image manipulation that learns to modify images based on instructions without needing target images during training.

Contribution

We propose a cyclic GAN model that enables weakly supervised, text-guided image editing by verifying semantic correctness and learning to undo modifications.

Findings

Effective on CLEVR and COCO datasets

Outperforms existing methods in semantic accuracy

Demonstrates strong generalization capabilities

Abstract

We tackle the problem of target-free text-guided image manipulation, which requires one to modify the input reference image based on the given text instruction, while no ground truth target image is observed during training. To address this challenging task, we propose a Cyclic-Manipulation GAN (cManiGAN) in this paper, which is able to realize where and how to edit the image regions of interest. Specifically, the image editor in cManiGAN learns to identify and complete the input image, while cross-modal interpreter and reasoner are deployed to verify the semantic correctness of the output image based on the input instruction. While the former utilizes factual/counterfactual description learning for authenticating the image semantics, the latter predicts the "undo" instruction and provides pixel-level supervision for the training of cManiGAN. With such operational cycle-consistency, our cManiGAN can be trained in the above weakly supervised setting. We conduct extensive experiments on the datasets of CLEVR and COCO, and the effectiveness and generalizability of our proposed method can be successfully verified. Project page: https://sites.google.com/view/wancyuanfan/projects/cmanigan.