Hierarchical Object Parsing from Structured Noisy Point Clouds

TL;DR

This paper presents a hierarchical Bayesian model and an efficient inference algorithm for object parsing from noisy structured point clouds, achieving state-of-the-art accuracy without relying on intensity data.

Contribution

It introduces a novel hierarchical Bayesian shape model with a data-driven inference method tailored for noisy point cloud data.

Findings

Achieves state-of-the-art parsing errors on standard datasets

Operates effectively without using intensity information

Handles structured noisy point clouds efficiently

Abstract

Object parsing and segmentation from point clouds are challenging tasks because the relevant data is available only as thin structures along object boundaries or other features, and is corrupted by large amounts of noise. To handle this kind of data, flexible shape models are desired that can accurately follow the object boundaries. Popular models such as Active Shape and Active Appearance models lack the necessary flexibility for this task, while recent approaches such as the Recursive Compositional Models make model simplifications in order to obtain computational guarantees. This paper investigates a hierarchical Bayesian model of shape and appearance in a generative setting. The input data is explained by an object parsing layer, which is a deformation of a hidden PCA shape model with Gaussian prior. The paper also introduces a novel efficient inference algorithm that uses informed data-driven proposals to initialize local searches for the hidden variables. Applied to the problem of object parsing from structured point clouds such as edge detection images, the proposed approach obtains state of the art parsing errors on two standard datasets without using any intensity information.