A Deep Learning Driven Active Framework for Segmentation of Large 3D Shape Collections

TL;DR

This paper introduces a deep learning-based active learning framework for efficiently segmenting large collections of 3D shapes, reducing user effort while maintaining high segmentation quality.

Contribution

It presents a novel active learning pipeline with a fast deep model, an information-theoretic shape selection method, and effective refinement tools for large-scale 3D shape segmentation.

Findings

Reduces user time and effort in shape segmentation

Achieves higher accuracy than state-of-the-art methods

Generalizes well across large datasets

Abstract

High-level shape understanding and technique evaluation on large repositories of 3D shapes often benefit from additional information known about the shapes. One example of such information is the semantic segmentation of a shape into functional or meaningful parts. Generating accurate segmentations with meaningful segment boundaries is, however, a costly process, typically requiring large amounts of user time to achieve high quality results. In this paper we present an active learning framework for large dataset segmentation, which iteratively provides the user with new predictions by training new models based on already segmented shapes. Our proposed pipeline consists of three novel components. First, we a propose a fast and relatively accurate feature-based deep learning model to provide dataset-wide segmentation predictions. Second, we propose an information theory measure to estimate the prediction quality and for ordering subsequent fast and meaningful shape selection. Our experiments show that such suggestive ordering helps reduce users time and effort, produce high quality predictions, and construct a model that generalizes well. Finally, we provide effective segmentation refinement features to help the user quickly correct any incorrect predictions. We show that our framework is more accurate and in general more efficient than state-of-the-art, for massive dataset segmentation with while also providing consistent segment boundaries.