Experimental Data for Folding-Based Point Cloud Attribute Compression

Experimental data from a 2020 IEEE ICIP paper on folding-based compression of point cloud attributes. The data includes trained models, compressed and decompressed patches, images, point clouds, and objective metrics for each patch and point cloud. It was provided by author Maurice Quach of Laboratoire des signaux et systèmes.

Use Cases

• Benchmarking compression algorithms based on provided objective metrics for patches and point clouds.
• Training or fine-tuning machine learning models for point cloud compression using the included trained models.
• Analyzing the visual quality of decompressed point clouds based on the provided images and point cloud files.
• Studying the performance of the folding-based compression method across different patches.

Strengths

• Includes trained machine learning models for the compression method described in the paper.
• Provides both compressed and decompressed data patches, images, and point clouds for direct comparison.
• Contains objective metrics calculated for each individual patch and each full point cloud.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Maurice Quach, Laboratoire des signaux et systèmes

Collection Method

Data generated for the experiments in the associated 2020 IEEE ICIP conference paper.