Supplementary Video Material for the Dissertation "Facial Attractiveness Enhancement in Videos"

Read full text
Authors: MAYCON SILVA
Cited by: 0

Abstract

This dataset contains processed video material used in the experimental evaluation of an automated facial attractiveness enhancement pipeline applied to video sequences.For each original video, two corresponding versions are provided. For every video identifier, the following files are included: *_original:The original, unprocessed video sequence, used as the baseline reference. *_enhanced:Fully processed video, including geometric transformation, perceptual refinement, texture enhancement, and GAN-based restoration. The videos were generated as part of the experimental analysis presented in the associated master’s dissertation and were used to evaluate three main aspects:(i) facial attractiveness enhancement,(ii) temporal stability across frames, and(iii) preservation of facial identity. This material is intended as supplementary content to support transparency, reproducibility, and visual inspection of the reported results. The dataset does not aim at biometric identification and is limited to experimental, research-oriented usage.

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
4 January 2026