Robust Pooling through the Data Mode

TL;DR

This paper introduces a novel robust pooling layer for point cloud neural networks that improves robustness against noise and outliers while maintaining real-time performance, by detecting data modes using RANSAC and histogram methods.

Contribution

It presents a new robust pooling layer that enhances neural network robustness to noisy data with faster computation compared to existing methods.

Findings

Enhanced robustness in point cloud classification and segmentation.

Faster processing speed than current robust approaches.

Effective in both point-based and graph-based neural network frameworks.

Abstract

The task of learning from point cloud data is always challenging due to the often occurrence of noise and outliers in the data. Such data inaccuracies can significantly influence the performance of state-of-the-art deep learning networks and their ability to classify or segment objects. While there are some robust deep learning approaches, they are computationally too expensive for real-time applications. This paper proposes a deep learning solution that includes a novel robust pooling layer which greatly enhances network robustness and performs significantly faster than state-of-the-art approaches. The proposed pooling layer looks for data a mode/cluster using two methods, RANSAC, and histogram, as clusters are indicative of models. We tested the pooling layer into frameworks such as Point-based and graph-based neural networks, and the tests showed enhanced robustness as compared to robust state-of-the-art methods.