SSLayout360: Semi-Supervised Indoor Layout Estimation from 360-Degree Panorama

TL;DR

This paper introduces SSLayout360, a semi-supervised method for indoor 3D layout estimation from 360-degree panoramas that reduces the need for labeled data while maintaining high accuracy.

Contribution

It is the first to learn room corner and boundary representations using both labeled and unlabeled data for panoramic scene layout estimation.

Findings

Achieves comparable accuracy to fully supervised methods with only 12% of labels.

Effective in complex indoor scenes with as few as 20 labeled examples.

Demonstrates significant reduction in labeled data requirement for 3D layout estimation.

Abstract

Recent years have seen flourishing research on both semi-supervised learning and 3D room layout reconstruction. In this work, we explore the intersection of these two fields to advance the research objective of enabling more accurate 3D indoor scene modeling with less labeled data. We propose the first approach to learn representations of room corners and boundaries by using a combination of labeled and unlabeled data for improved layout estimation in a 360-degree panoramic scene. Through extensive comparative experiments, we demonstrate that our approach can advance layout estimation of complex indoor scenes using as few as 20 labeled examples. When coupled with a layout predictor pre-trained on synthetic data, our semi-supervised method matches the fully supervised counterpart using only 12% of the labels. Our work takes an important first step towards robust semi-supervised layout estimation that can enable many applications in 3D perception with limited labeled data.