Semantic Line Combination Detector

TL;DR

The paper introduces SLCD, a new algorithm that optimally combines semantic lines to improve detection accuracy and applicability across multiple vision tasks such as vanishing point detection, symmetry axis detection, and image retrieval.

Contribution

It presents a novel semantic line combination detector that processes all lines simultaneously for better harmony assessment, outperforming existing methods.

Findings

SLCD outperforms existing semantic line detectors on various datasets.

SLCD effectively applies to vanishing point detection, symmetry axis detection, and image retrieval.

The code implementation is publicly available at GitHub.

Abstract

A novel algorithm, called semantic line combination detector (SLCD), to find an optimal combination of semantic lines is proposed in this paper. It processes all lines in each line combination at once to assess the overall harmony of the lines. First, we generate various line combinations from reliable lines. Second, we estimate the score of each line combination and determine the best one. Experimental results demonstrate that the proposed SLCD outperforms existing semantic line detectors on various datasets. Moreover, it is shown that SLCD can be applied effectively to three vision tasks of vanishing point detection, symmetry axis detection, and composition-based image retrieval. Our codes are available at https://github.com/Jinwon-Ko/SLCD.