polars-cv

High-performance vision and array processing for Polars DataFrames.

polars-cv is a Polars plugin that enables lazy, modular image processing on DataFrame columns.

Key Features

• 🚀 Performance: Rust-powered operations with automatic kernel fusion.
• 🔗 Composable: Modular pipelines that can be branched and merged.
• 🎯 Multi-Domain: Seamlessly move between images, geometry (contours), and numeric results.
• 📊 Multi-Output: Extract multiple results from a single execution pass.
• 🔌 Integration: Direct output to NumPy and other formats.

Quick Example

import polars as pl
from polars_cv import Pipeline

# Define a reusable preprocessing pipeline
preprocess = (
    Pipeline()
    .source("image_bytes")
    .resize(height=224, width=224)
    .grayscale()
)

# Apply to a DataFrame column and output as NumPy bytes
df = pl.DataFrame({"image": [image_bytes]})
result = df.with_columns(
    processed=pl.col("image").cv.pipe(preprocess).sink("numpy")
)

Getting Started

1. Installation
2. Quickstart
3. User Guide