Human-centric Indoor Scene Synthesis Using Stochastic Grammar
This work addresses the need for large-scale, high-quality synthetic data for computer vision tasks, offering an incremental improvement over existing scene synthesis methods.
The paper tackles the problem of generating realistic indoor scenes by proposing a human-centric method using a stochastic grammar model to synthesize 3D room layouts and 2D images, achieving robust sampling with high visual realism and accuracy in affordance maps as evaluated by human subjects.
We present a human-centric method to sample and synthesize 3D room layouts and 2D images thereof, to obtain large-scale 2D/3D image data with perfect per-pixel ground truth. An attributed spatial And-Or graph (S-AOG) is proposed to represent indoor scenes. The S-AOG is a probabilistic grammar model, in which the terminal nodes are object entities. Human contexts as contextual relations are encoded by Markov Random Fields (MRF) on the terminal nodes. We learn the distributions from an indoor scene dataset and sample new layouts using Monte Carlo Markov Chain. Experiments demonstrate that our method can robustly sample a large variety of realistic room layouts based on three criteria: (i) visual realism comparing to a state-of-the-art room arrangement method, (ii) accuracy of the affordance maps with respect to groundtruth, and (ii) the functionality and naturalness of synthesized rooms evaluated by human subjects. The code is available at https://github.com/SiyuanQi/human-centric-scene-synthesis.