SynTable: A Synthetic Data Generation Pipeline for Unseen Object Amodal Instance Segmentation of Cluttered Tabletop Scenes

arXiv:2307.07333v2 Announce Type: replace
Abstract: In this work, we present SynTable, a unified and flexible Python-based dataset generator built using NVIDIA's Isaac Sim Replicator Composer for generating high-quality synthetic datasets for unseen object amodal instance segmentation of cluttered tabletop scenes. Our dataset generation tool can render a complex 3D scene containing object meshes, materials, textures, lighting, and backgrounds. Metadata, such as modal and amodal instance segmentation masks, occlusion masks, depth maps, bounding boxes, and material properties, can be generated to …

abstract arxiv cs.cv data dataset dataset generation datasets generator instance isaac sim nvidia pipeline python quality segmentation sim synthetic synthetic data tool type work