PartNet: A Large-Scale Benchmark for Fine-Grained and Hierarchical Part-Level 3D Object Understanding

Abstract

We present PartNet: a consistent, large-scale dataset of 3D objects annotated with fine-grained, instance-level, and hierarchical 3D part information. Our dataset consists of 573,585 part instances over 26,671 3D models covering 24 object categories. This dataset enables and serves as a catalyst for many tasks such as shape analysis, dynamic 3D scene modeling and simulation, affordance analysis, and others. Using our dataset, we establish three benchmarking tasks for evaluating 3D part recognition: fine-grained semantic segmentation, hierarchical semantic segmentation, and instance segmentation. We benchmark four state-ofthe-art 3D deep learning algorithms for fine-grained semantic segmentation and three baseline methods for hierarchical semantic segmentation. We also propose a novel method for part instance segmentation and demonstrate its superior performance over existing methods.

Towards a Framework for Visual Intelligence in Service Robotics: Epistemic Requirements and Gap Analysis

“…Albeit including a lower number of models than ShapeNetCore, ShapeNetSem is augmented with richer annotations describing the physical properties of objects, e.g., absolute size estimations, upright and front orientation and others (Savva et al, 2015). Most recently, "part of" annotations for a subset of ShapeNet models spanning across 24 object categories were released as PartNet (Mo et al, 2019). Lake et al (2017) have recently suggested a set of core ingredients that characterise the way we think and learn.…”

PartNet: A Recursive Part Decomposition Network for Fine-Grained and Hierarchical Shape Segmentation

“…With the advances in deep learning based 3D shape seg-mentation, a benchmark for instance segmentation of finegrained parts is called for. A nice benchmark for evaluating fine-grained shape segmentation is recently proposed in a concurrent work in [16]. In this work, we propose FineSeg.…”

A survey on deep geometry learning: From a representation perspective

“…The whole dataset includes 49,884 valid floors with 404,058 rooms and 5,697,217 object instances. PartNet [114] provides a more detailed CAD model dataset with fine-grained, hierarchical part annotations, bringing more challenges, and resources for 3D object applications such as semantic segmentation, shape editing, and shape generation. 3D-Future [115] provides a large-scale furniture dataset, which includes over 20,000 scenes in over 5000 rooms with over 10,000 3D instances.…”