MMPD-Bench: Mueller Matrix Polarimetric Decomposition Benchmark

MMPD-Bench is a benchmark for learning mappings from Mueller matrix observations to polarimetric decomposition modalities. Each sample contains a channel-first Mueller matrix tensor and a target tensor with six Lu-Chipman reference modalities. The dataset, authored by HY2333, includes external waveplate test data at 633 nm and spectral test data at 610, 650, and 690 nm, with a last update recorded on 2026-05-22.

Use Cases

• Training models to predict Lu-Chipman decomposition modalities based on Mueller matrix observations.
• Benchmarking algorithm performance for mapping polarimetric imaging data to reference modalities.
• Developing spectral analysis models using data from 610, 650, and 690 nm wavelengths.
• Validating polarimetric imaging systems with external waveplate test data at 633 nm.

Strengths

• Includes spectral test data at three distinct wavelengths (610, 650, and 690 nm).
• Provides external waveplate test data at a specific wavelength (633 nm).
• Each sample contains a target tensor with six defined Lu-Chipman reference modalities.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Hugging Face dataset authored by HY2333.

Collection Method

Likely contains experimentally or synthetically generated polarimetric imaging data.

Time Range

null

Freshness

Last updated 2026-05-22 17:04:11; freshness should be verified.

Geography

null