Recurrence-based Vanishing Point Detection

TL;DR

This paper introduces an unsupervised recurrence-based method for vanishing point detection that leverages implicit recurring patterns, outperforming classical methods and matching supervised approaches on real-world data.

Contribution

The paper presents a novel unsupervised recurrence-based approach for vanishing point detection and provides two new datasets for evaluation.

Findings

Outperforms classical methods on synthetic data

Matches supervised methods on real-world data

Introduces two new vanishing point datasets

Abstract

Classical approaches to Vanishing Point Detection (VPD) rely solely on the presence of explicit straight lines in images, while recent supervised deep learning approaches need labeled datasets for training. We propose an alternative unsupervised approach: Recurrence-based Vanishing Point Detection (R-VPD) that uses implicit lines discovered from recurring correspondences in addition to explicit lines. Furthermore, we contribute two Recurring-Pattern-for-Vanishing-Point (RPVP) datasets: 1) a Synthetic Image dataset with 3,200 ground truth vanishing points and camera parameters, and 2) a Real-World Image dataset with 1,400 human annotated vanishing points. We compare our method with two classical methods and two state-of-the-art deep learning-based VPD methods. We demonstrate that our unsupervised approach outperforms all the methods on the synthetic images dataset, outperforms the classical methods, and is on par with the supervised learning approaches on real-world images.