# Corners for Layout: End-to-End Layout Recovery from 360 Images

## TL;DR

This paper introduces CFL, an end-to-end model for 3D indoor layout recovery from 360 images that relaxes scene assumptions, improves generalization, and operates efficiently for real-time applications.

## Contribution

CFL is the first end-to-end model for 3D layout recovery from 360 images that outperforms existing methods while being computationally efficient and less assumption-dependent.

## Key findings

- Outperforms state-of-the-art methods in layout recovery.
- Generalizes better to camera position variations.
- Operates efficiently for real-time applications.

## Abstract

The problem of 3D layout recovery in indoor scenes has been a core research topic for over a decade. However, there are still several major challenges that remain unsolved. Among the most relevant ones, a major part of the state-of-the-art methods make implicit or explicit assumptions on the scenes -- e.g. box-shaped or Manhattan layouts. Also, current methods are computationally expensive and not suitable for real-time applications like robot navigation and AR/VR. In this work we present CFL (Corners for Layout), the first end-to-end model for 3D layout recovery on 360 images. Our experimental results show that we outperform the state of the art relaxing assumptions about the scene and at a lower cost. We also show that our model generalizes better to camera position variations than conventional approaches by using EquiConvs, a type of convolution applied directly on the sphere projection and hence invariant to the equirectangular distortions. CFL Webpage: https://cfernandezlab.github.io/CFL/