Doppelgangers: Learning to Disambiguate Images of Similar Structures

TL;DR

This paper introduces a learning-based method and a new dataset for disambiguating visually similar images of different 3D surfaces, improving 3D reconstruction accuracy.

Contribution

The authors present a novel dataset and a network architecture for disambiguating similar images, enhancing 3D reconstruction reliability.

Findings

The method effectively distinguishes illusory matches in challenging cases.

Integration into SfM pipelines improves 3D reconstruction correctness.

The dataset facilitates training and benchmarking for disambiguation tasks.

Abstract

We consider the visual disambiguation task of determining whether a pair of visually similar images depict the same or distinct 3D surfaces (e.g., the same or opposite sides of a symmetric building). Illusory image matches, where two images observe distinct but visually similar 3D surfaces, can be challenging for humans to differentiate, and can also lead 3D reconstruction algorithms to produce erroneous results. We propose a learning-based approach to visual disambiguation, formulating it as a binary classification task on image pairs. To that end, we introduce a new dataset for this problem, Doppelgangers, which includes image pairs of similar structures with ground truth labels. We also design a network architecture that takes the spatial distribution of local keypoints and matches as input, allowing for better reasoning about both local and global cues. Our evaluation shows that our method can distinguish illusory matches in difficult cases, and can be integrated into SfM pipelines to produce correct, disambiguated 3D reconstructions. See our project page for our code, datasets, and more results: http://doppelgangers-3d.github.io/.