The Replica Dataset: A Digital Replica of Indoor Spaces

TL;DR

Replica is a comprehensive dataset of highly realistic 3D indoor scenes designed to advance machine learning research in visual perception, semantic understanding, and embodied agent navigation, with potential for real-world transfer.

Contribution

The paper introduces Replica, a new large-scale, highly realistic 3D indoor dataset with detailed semantic and reflectance information, supporting diverse ML applications and compatible with Habitat.

Findings

Provides 18 photorealistic 3D indoor scenes with detailed annotations

Enables training of ML models for vision, segmentation, and navigation tasks

Supports transfer learning to real-world environments

Abstract

We introduce Replica, a dataset of 18 highly photo-realistic 3D indoor scene reconstructions at room and building scale. Each scene consists of a dense mesh, high-resolution high-dynamic-range (HDR) textures, per-primitive semantic class and instance information, and planar mirror and glass reflectors. The goal of Replica is to enable machine learning (ML) research that relies on visually, geometrically, and semantically realistic generative models of the world - for instance, egocentric computer vision, semantic segmentation in 2D and 3D, geometric inference, and the development of embodied agents (virtual robots) performing navigation, instruction following, and question answering. Due to the high level of realism of the renderings from Replica, there is hope that ML systems trained on Replica may transfer directly to real world image and video data. Together with the data, we are releasing a minimal C++ SDK as a starting point for working with the Replica dataset. In addition, Replica is `Habitat-compatible', i.e. can be natively used with AI Habitat for training and testing embodied agents.