The farthest point strategy for progressive image sampling

1997/09/01 by Y. Eldar, Michael Lindenbaum, M. Lindenbaum +3 · 31 citations
Computer Science · Engineering · #Advanced Image Processing Techniques #Medical Image Segmentation Techniques #Sparse and Compressive Sensing Techniques

paper · doi:10.1109/83.623193

openalex publication_date 1997/09/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/29

Abstract

A new method of farthest point strategy (FPS) for progressive image acquisition-an acquisition process that enables an approximation of the whole image at each sampling stage-is presented. Its main advantage is in retaining its uniformity with the increased density, providing efficient means for sparse image sampling and display. In contrast to previously presented stochastic approaches, the FPS guarantees the uniformity in a deterministic min-max sense. Within this uniformity criterion, the sampling points are irregularly spaced, exhibiting anti-aliasing properties comparable to those characteristic of the best available method (Poisson disk). A straightforward modification of the FPS yields an image-dependent adaptive sampling scheme. An efficient O(N log N) algorithm for both versions is introduced, and several applications of the FPS are discussed.

Cited by

• Deep Magnification-Flexible Upsampling over 3D Point Clouds
• Sign-Agnostic Implicit Learning of Surface Self-Similarities for Shape Modeling and Reconstruction from Raw Point Clouds
• Vote from the Center: 6 DoF Pose Estimation in RGB-D Images by Radial Keypoint Voting
• Gaussian Process Regression for Materials and Molecules
• PCRNet: Point Cloud Registration Network using PointNet Encoding
• Hydrogen atom scattering at the Al 2 O 3 (0001) surface: a combined experimental and theoretical study
• A Point Set Generation Network for 3D Object Reconstruction from a Single Image
• Exploiting Hierarchical Interactions for Protein Surface Learning
• 3D Point Cloud Denoising using Graph Laplacian Regularization of a Low Dimensional Manifold Model
• MOPS-Net: A Matrix Optimization-driven Network forTask-Oriented 3D Point Cloud Downsampling
• Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetry
• 6D Object Pose Regression via Supervised Learning on Point Clouds
• SK-Net: Deep Learning on Point Cloud via End-to-end Discovery of Spatial Keypoints
• DS++: A flexible, scalable and provably tight relaxation for matching problems
• Training Data Set Refinement for the Machine Learning Potential of Li-Si Alloys via Structural Similarity Analysis
• Sampling theory for spatial field sensing: Application to electro- and magnetoencephalography
• Spatial sampling of MEG and EEG revisited: From spatial-frequency spectra to model-informed sampling
• One Framework to Register Them All: PointNet Encoding for Point Cloud Alignment
• Farthest Point Sampling in Property Designated Chemical Feature Space as a General Strategy for Enhancing the Machine Learning Model Performance for Small Scale Chemical Dataset
• Learning Scene Dynamics from Point Cloud Sequences
• PointHop++: A Lightweight Learning Model on Point Sets for 3D Classification
• The continuous Procrustes distance between two surfaces
• PointTrackNet: An End-to-End Network For 3-D Object Detection and Tracking From Point Clouds
• Detection and Characterization of Intrinsic Symmetry
• Pedestrian Detection in 3D Point Clouds using Deep Neural Networks
• DecoSurf: Recursive Geodesic Patterns on Triangle Meshes
• Label-Free Model Evaluation with Semi-Structured Dataset Representations
• NPTC-net: Narrow-Band Parallel Transport Convolutional Neural Network on Point Clouds
• Curvature-Variation-Inspired Sampling for Point Cloud Classification and Segmentation
• Towards Fine-grained 3D Face Dense Registration: An Optimal Dividing and Diffusing Method
• Bregman Voronoi Diagrams

Related

• The General sampling theorem, Compressed sensing and a method of image sampling and reconstruction with sampling rates close to the theoretical limit
• Progressive versus Random Projections for Compressive Capture of Images, Lightfields and Higher Dimensional Visual Signals
• Improving spatial domain based image formation through compressed sensing
• Measurement-Adaptive Sparse Image Sampling and Recovery
• Forget Superresolution, Sample Adaptively (when Path Tracing)