Automatically Annotating Indoor Images with CAD Models via RGB-D Scans

TL;DR

This paper introduces an automatic method for annotating indoor scene images with CAD models using RGB-D scans, achieving accuracy comparable to manual annotations and enabling large-scale dataset annotation.

Contribution

The method employs an analysis-by-synthesis approach with a novel cloning procedure to automatically annotate objects with CAD models in indoor scenes.

Findings

Annotations are as accurate as manual ones according to expert evaluation.

The approach successfully annotates large datasets like ScanNet and ARKitScenes.

Enables automatic, large-scale annotation of indoor scene datasets.

Abstract

We present an automatic method for annotating images of indoor scenes with the CAD models of the objects by relying on RGB-D scans. Through a visual evaluation by 3D experts, we show that our method retrieves annotations that are at least as accurate as manual annotations, and can thus be used as ground truth without the burden of manually annotating 3D data. We do this using an analysis-by-synthesis approach, which compares renderings of the CAD models with the captured scene. We introduce a 'cloning procedure' that identifies objects that have the same geometry, to annotate these objects with the same CAD models. This allows us to obtain complete annotations for the ScanNet dataset and the recent ARKitScenes dataset.