Mask Guided Gated Convolution for Amodal Content Completion

We present a model to reconstruct partially visible objects. The model takes a mask as an input, which we call weighted mask. The mask is utilized by gated convolutions to assign more weight to the visible pixels of the occluded instance compared to the background, while ignoring the features of the invisible pixels. By drawing more attention from the visible region, our model can predict the invisible patch more effectively than the baseline models, especially in instances with uniform texture. The model is trained on COCOA dataset and two subsets of it in a self-supervised manner. The results demonstrate that our model generates higher quality and more texture-rich outputs compared to baseline models. Code is available at: https://github.com/KaziwaSaleh/mask-guided

I. Introduction

Our environment is cluttered and crowded. Objects are often occluded by other object(s), and occlusion can happen in various rates, and positions. Regardless of the fact that we encounter more occluded objects than fully visible ones, we seamlessly and effortlessly can imagine the full shape and gestalt of partially visible objects. This process is called amodal completion. While this task is natural and innate for humans, it is challenging for machines. However, it is essential for many real-world computer vision based applications, such as semantic scene understanding, robotics, autonomous vehicles, and security and surveillance systems. [1].