InteriorNet: Mega-scale Multi-sensor Photo-realistic Indoor Scenes Dataset
This dataset addresses the need for scalable, realistic synthetic data in computer vision, particularly for SLAM algorithms, though it is incremental as it builds on existing synthetic dataset efforts.
The authors introduced InteriorNet, a large-scale synthetic indoor scenes dataset designed to offer higher photo-realism, scale, and variability for computer vision tasks like SLAM benchmarking, by rendering video sequences from professional interior designs and furniture assets.
Datasets have gained an enormous amount of popularity in the computer vision community, from training and evaluation of Deep Learning-based methods to benchmarking Simultaneous Localization and Mapping (SLAM). Without a doubt, synthetic imagery bears a vast potential due to scalability in terms of amounts of data obtainable without tedious manual ground truth annotations or measurements. Here, we present a dataset with the aim of providing a higher degree of photo-realism, larger scale, more variability as well as serving a wider range of purposes compared to existing datasets. Our dataset leverages the availability of millions of professional interior designs and millions of production-level furniture and object assets -- all coming with fine geometric details and high-resolution texture. We render high-resolution and high frame-rate video sequences following realistic trajectories while supporting various camera types as well as providing inertial measurements. Together with the release of the dataset, we will make executable program of our interactive simulator software as well as our renderer available at https://interiornetdataset.github.io. To showcase the usability and uniqueness of our dataset, we show benchmarking results of both sparse and dense SLAM algorithms.