PyImageCUDA

PyImageCUDA is a specialized library for creative image composition rather than computer vision or analysis. It provides GPU-accelerated effects for design workflows—operations typically found in professional design software.

Built for composition, not analysis - Focused on creative workflows, not edge detection or object recognition.

Quick Start

from pyimagecuda import Image, Fill, Effect, Blend, Transform, save

with Image(1024, 1024) as bg:
    Fill.color(bg, (0, 1, 0.8, 1))
    with Image(512, 512) as card:
        Fill.gradient(card, (1, 0, 0, 1), (0, 0, 1, 1), 'radial')
        Effect.rounded_corners(card, 50)

        with Effect.stroke(card, 10, (1, 1, 1, 1)) as stroked:
            with Effect.drop_shadow(stroked,
                                    blur=50,
                                    color=(0, 0, 0, 1)) as shadowed:
                with Transform.rotate(shadowed, 45) as rotated:
                    Blend.normal(bg, rotated, anchor='center')

    save(bg, 'output.png')

Installation

pip install pyimagecuda

Requirements:

Windows 10/11 (64-bit), Linux coming soon
NVIDIA GPU (GTX 900 series or newer)
Standard NVIDIA drivers

No CUDA Toolkit or Visual Studio required.

Core Concepts

Images and Memory

Two image types: Image (float32) for processing, ImageU8 (uint8) for I/O. Memory management: automatic (GC), explicit (with), or manual (.free()).

→ Full details: Image & Memory