Awesome

Illumination Histogram Consistency Metric for Quantitative Assessment of Video Sequences

This is the official implementation of our IHC metric for quantitative assessment of video sequences

Dependencies

Matlab Python

Data Download

We used 3D software Blender https://www.blender.org/ to generate 100 continuous video frames with linearly increasing illumination. The simulated video frames can be downloaded from https://1drv.ms/u/s!At3lDmLw-VwA2iLMCDxrk5miWksj?e=Cz5TYF

Illumination Map Generation using the Retinex Model

Run the SSR algorithm using Matlab to produce the illumination maps

SSR_Code.m

Histogram Geneation and IHC Computation

Run

python IHC_histogram.py

Citation

This paper is under review now. If you use this code, please cite our paper later:

@InProceedings{IHC_2024,
    author    = {Long Chen, Mobarakol Islam, Matt Clarkson, and Thomas Dowrick},
    title     = {Illumination Histogram Consistency Metric for Quantitative Assessment of Video Sequences},
}

License

Licensed under the CC BY-NC-SA 4.0 license (https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode). Copyright (C) 2024 NVIDIA Corporation. All rights reserved. Licensed under the CC BY-NC-SA 4.0 license (https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode). These licenses allow you to use, modify and share the project for non-commercial use as long as you adhere to the conditions of the license above.

Contact

If you have any questions, do not hesitate to contact us: chenlongcv@gmail.com