Profile Encode Repository

Owned by R D
Last updated: 2024-07-27 by Maxwell Lou
1 min read

Overview

The time and space performance of different image encoding formats is measured to support image transmission.

Repository:

GitHub - UWARG/profile-encode: Profile usage of time and space of various image encoders

Software

Setup

Follow the instructions to clone the repository and activate the environment: Autonomy Workflow Software

Install packages:

pip install -r requirements.txt

Download and extract the test dataset:

  1. In the repository root, create a new directory: test_images

  2. Open the WARG OneDrive: Autonomy OneDrive

  3. Download: Encode Test Dataset 2024

  4. Extract the contents so that the images are in test_images/Encode Test Dataset 2024 (e.g. test_images/Encode Test Dataset 2024/0.png )

Usage

Activate the environment: Autonomy Workflow Software

Run any of the benchmarks:

python -m avif_benchmark python -m heif_benchmark python -m png_benchmark python -m jpeg_benchmark

Each benchmark takes around 1-2 hours to complete on the WARG desktop.

Record the results:

Hardware

Performance is to be measured on the hardware running image transmission.

Raspberry Pi: Raspberry Pi

Â