Pixelator Operator

Authors: NVIDIA Holoscan Team (NVIDIA)
Supported platforms: x86_64, aarch64
Language: Python
Last modified: May 22, 2025
Latest version: 1.0
Minimum Holoscan SDK version: 3.0.0
Tested Holoscan SDK versions: 3.1.0
Contribution metric: Level 1 - Highly Reliable

A Pixelation-based Deidentification Operator

Overview

In medical and sensitive imaging workflows, pixelation is a common method for deidentification. The PixelatorOp is a Holoscan operator that performs pixelation-based deidentification on input images, suitable for applications such as surgical video anonymization where the camera may get out of the body and capture sensitive or protected information.

Requirements

• Holoscan SDK
• cupy

Example Usage

from holohub.operators.deidentification.pixelator import PixelatorOp
op = PixelatorOp(block_size_h=16, block_size_w=16)

Parameters

• tensor_name: The name of the tensor to be pixelated.
• block_size_h: Height of the pixelation block.
• block_size_w: Width of the pixelation block.