Image Completion and Extrapolation with Contextual Cycle Consistency

TL;DR

This paper introduces a novel training technique that enables a neural network to perform both image completion and extrapolation effectively by training both tasks concurrently, improving generalization across different image editing scenarios.

Contribution

The paper proposes a joint training method for image completion and extrapolation networks, enhancing their ability to generalize and perform well on both tasks simultaneously.

Findings

Efficient completion of large missing regions.

Improved generalization for both tasks.

Competitive results against state-of-the-art methods.

Abstract

Image Completion refers to the task of filling in the missing regions of an image and Image Extrapolation refers to the task of extending an image at its boundaries while keeping it coherent. Many recent works based on GAN have shown progress in addressing these problem statements but lack adaptability for these two cases, i.e. the neural network trained for the completion of interior masked images does not generalize well for extrapolating over the boundaries and vice-versa. In this paper, we present a technique to train both completion and extrapolation networks concurrently while benefiting each other. We demonstrate our method's efficiency in completing large missing regions and we show the comparisons with the contemporary state of the art baseline.