Amodal Ground Truth and Completion in the Wild

TL;DR

This paper introduces an automatic pipeline using 3D data to generate authentic amodal segmentation ground truth for occluded objects in real images, creating a new benchmark and improving state-of-the-art performance.

Contribution

It presents a novel automatic method for amodal ground truth generation and a new benchmark dataset, advancing amodal segmentation in real-world scenarios.

Findings

Achieved state-of-the-art results on amodal segmentation datasets.

Developed two architecture variants for amodal completion.

Created MP3D-Amodal, a new diverse amodal segmentation benchmark.

Abstract

This paper studies amodal image segmentation: predicting entire object segmentation masks including both visible and invisible (occluded) parts. In previous work, the amodal segmentation ground truth on real images is usually predicted by manual annotaton and thus is subjective. In contrast, we use 3D data to establish an automatic pipeline to determine authentic ground truth amodal masks for partially occluded objects in real images. This pipeline is used to construct an amodal completion evaluation benchmark, MP3D-Amodal, consisting of a variety of object categories and labels. To better handle the amodal completion task in the wild, we explore two architecture variants: a two-stage model that first infers the occluder, followed by amodal mask completion; and a one-stage model that exploits the representation power of Stable Diffusion for amodal segmentation across many categories. Without bells and whistles, our method achieves a new state-of-the-art performance on Amodal segmentation datasets that cover a large variety of objects, including COCOA and our new MP3D-Amodal dataset. The dataset, model, and code are available at https://www.robots.ox.ac.uk/~vgg/research/amodal/.