Mendevi 1.2.4

Measures of Encoding and Decoding of Videos.

[license GPL] [linting: pylint] [versions] [downloads] [documentation]

Useful links: Binary Installers | Source Repository | Online Documentation |

Description

This module performs energy and metrics measurements on video, for encoding and decoding. It also provides a detailed dataset.

It manages the following parameters:

  1. It supports the libx264, libx265, libvpx-vp9, libsvtav1, librav1e and vvc encoders.

  2. Distortions are measured using the lpips, psnr, ssim and vmaf metrics.

  3. Complexity are measured using the rms_sobel and rms_time_diff metrics.

  4. Encoding efforts are fast, medium and slow.

  5. It takes care about the colorspaces.

  6. Iterate over different effort, encoder, mode, quality, threads, fps, resolution and pix_fmt.

  7. Energy measurements are catched with RAPL and an external wattmeter on grid'5000.

  8. Get the cpu and ram activity.

  9. Get a full environment context, including harware and software version.

  10. It support the mode (constant bitrate) cbr and (constant quality) vbr.

Pipeline

This is the pipeline used for measurements:

Pipeline diagram

Example of result

Example of rate distortion curve:

mendevi plot mendevi.db -x bitrate -y psnr -y ssim -wx profile -c encoder
Result plot of rate distortion

Example of energy per encoder:

mendevi plot mendevi.db -x quality -y energy -wx profile -wy mode -c encoder -m effort
Result plot of encoding energy

Alternatives

  1. The GREEM video encoding measurement tool.

  2. The MVCD database also includes video encoding and decoding energy measurements.

  3. The COCONUT database also includes video decoding measurements.

  4. The SEED and VEED dataset offers a comprehensive LCA and GPU measurements.

  5. The CTC videos are used for the tests.

Getting started

Dataset

Developer guide

API documentation

Reference

The documentation was generated on 2025-12-01 12H37.