Physically-Based Rendering for Indoor Scene Understanding Using Convolutional Neural Networks

TL;DR

This paper introduces a large-scale synthetic dataset of physically-based rendered indoor scenes to improve training for scene understanding tasks, demonstrating that realistic rendering enhances model performance.

Contribution

It provides a comprehensive synthetic dataset and a systematic analysis of rendering effects, advancing training practices for indoor scene understanding with deep neural networks.

Findings

Realistic rendering improves training effectiveness.

Pretraining on the synthetic dataset outperforms existing methods.

Insights into optimal synthetic data generation for scene understanding.

Abstract

Indoor scene understanding is central to applications such as robot navigation and human companion assistance. Over the last years, data-driven deep neural networks have outperformed many traditional approaches thanks to their representation learning capabilities. One of the bottlenecks in training for better representations is the amount of available per-pixel ground truth data that is required for core scene understanding tasks such as semantic segmentation, normal prediction, and object edge detection. To address this problem, a number of works proposed using synthetic data. However, a systematic study of how such synthetic data is generated is missing. In this work, we introduce a large-scale synthetic dataset with 400K physically-based rendered images from 45K realistic 3D indoor scenes. We study the effects of rendering methods and scene lighting on training for three computer vision tasks: surface normal prediction, semantic segmentation, and object boundary detection. This study provides insights into the best practices for training with synthetic data (more realistic rendering is worth it) and shows that pretraining with our new synthetic dataset can improve results beyond the current state of the art on all three tasks.