A Comprehensive Performance Evaluation of 3D Local Feature Descriptors

By Yulan Guo, Mohammed Bennamoun, Ferdous Sohel, Min Lu, Jianwei Wan, N. Kwok. International Journal of Computer Vision, 2015

578 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/16b4a666e3369e1809a09d3b8d3c3d82ed9607c7

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

RPM-Net: Robust Point Matching Using Learned Features (2020, background): “A large variety of handcrafted 3D feature descriptors have been proposed and a comprehensive survey can be found in [13].”
Image Matching from Handcrafted to Deep Features: A Survey (2020, background): “Guo et al. (2016) divided the handcrafted descriptors into spatial distribution histogram- and geometric attribute histogram-based descriptors, with the former representing the local feature by histograms that encode spatial distributions of the points in the support region.”
D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features (2020, background): “Early approaches to extract 3D local descriptors are mainly hand-crafted [12], which generally lack robustness against noise and occlusion.”
PPFNet: Global Context Aware Local Features for Robust 3D Point Matching (2018, background): “Unfortunately, contrary to 2D, the repeatability and distinctiveness of 3D features were found to be way below expectations [10, 11, 24].”
Deep Learning Advances in Computer Vision with 3D Data (2017, methodology): “Moreover, a thorough investigation on the performance of several 3D local descriptors in the context of 3D object recognition, 3D modeling, and 3D shape retrieval is available in Guo et al. [2016a].”
3DmFV: Three-Dimensional Point Cloud Classification in Real-Time Using Convolutional Neural Networks (2018, background): “While the comprehensive performance evaluation in [17] suggests guidelines for feature selection, it remains non-trivial and highly data specific.”
Fully Convolutional Geometric Features (2019, background): “Guo et al. [13] provide a comprehensive review of such hand-crafted descriptors.”
Learning to Detect Good 3D Keypoints (2017, result): “5 in Guo et al. (2016)] It is worth noticing how, in the results dealing with matching SHOT descriptors reported in Fig.”
SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registration (2020, background): “the pipeline, fundamentally limiting the representational power of the framework [25].”
Fast Global Registration (2016, methodology): “We have chosen this feature because it can be computed in a fraction of a millisecond and provides good matching accuracy across a broad range of datasets [16].”
The Perfect Match: 3D Point Cloud Matching With Smoothed Densities (2018, background): “In fact, they still fail to handle point cloud resolution changes, noisy data, occlusions and clutter [10].”
T-LESS: An RGB-D Dataset for 6D Pose Estimation of Texture-Less Objects (2017, background): “Instead, recent approaches that can deal with texture-less objects have focused on local 3D feature description [33, 51, 19], and semi-global or 2017 IEEE Winter Conference on Applications of Computer Vision”