Three-Filters-to-Normal: An Accurate and Ultrafast Surface Normal Estimator

TL;DR

This paper introduces 3F2N, a simple yet highly accurate and ultrafast surface normal estimator for structured range sensor data, utilizing three filtering operations to outperform existing methods in accuracy and speed.

Contribution

The paper presents a novel surface normal estimation method using only three filters, achieving state-of-the-art accuracy and real-time processing speeds, with publicly available datasets and code.

Findings

Outperforms existing geometry-based SNEs in accuracy.

Achieves over 260 Hz in CPU and 21 kHz in GPU processing speeds.

Demonstrates robustness across easy, medium, and hard synthetic datasets.

Abstract

This paper proposes three-filters-to-normal (3F2N), an accurate and ultrafast surface normal estimator (SNE), which is designed for structured range sensor data, e.g., depth/disparity images. 3F2N SNE computes surface normals by simply performing three filtering operations (two image gradient filters in horizontal and vertical directions, respectively, and a mean/median filter) on an inverse depth image or a disparity image. Despite the simplicity of 3F2N SNE, no similar method already exists in the literature. To evaluate the performance of our proposed SNE, we created three large-scale synthetic datasets (easy, medium and hard) using 24 3D mesh models, each of which is used to generate 1800--2500 pairs of depth images (resolution: 480X640 pixels) and the corresponding ground-truth surface normal maps from different views. 3F2N SNE demonstrates the state-of-the-art performance, outperforming all other existing geometry-based SNEs, where the average angular errors with respect to the easy, medium and hard datasets are 1.66 degrees, 5.69 degrees and 15.31 degrees, respectively. Furthermore, our C++ and CUDA implementations achieve a processing speed of over 260 Hz and 21 kHz, respectively. Our datasets and source code are publicly available at sites.google.com/view/3f2n.