Floating No More: Object-Ground Reconstruction from a Single Image

TL;DR

This paper introduces ORG, a method for reconstructing 3D objects along with ground surfaces from a single image, improving the realism of object placement and interactions in 3D-aware editing tasks.

Contribution

The paper proposes a novel approach that jointly reconstructs object and ground geometry using pixel-level representations, addressing limitations of previous methods that ignore object-ground relationships.

Findings

Effective reconstruction of object-ground geometry on unseen data

Significant improvement in shadow rendering quality

Enhanced object pose manipulation capabilities

Abstract

Recent advancements in 3D object reconstruction from single images have primarily focused on improving the accuracy of object shapes. Yet, these techniques often fail to accurately capture the inter-relation between the object, ground, and camera. As a result, the reconstructed objects often appear floating or tilted when placed on flat surfaces. This limitation significantly affects 3D-aware image editing applications like shadow rendering and object pose manipulation. To address this issue, we introduce ORG (Object Reconstruction with Ground), a novel task aimed at reconstructing 3D object geometry in conjunction with the ground surface. Our method uses two compact pixel-level representations to depict the relationship between camera, object, and ground. Experiments show that the proposed ORG model can effectively reconstruct object-ground geometry on unseen data, significantly enhancing the quality of shadow generation and pose manipulation compared to conventional single-image 3D reconstruction techniques.