turian / dover-video-quality-assessment

DOVER video quality assessment tool, assigning videos both aesthetic and technical quality scores

  • Public
  • 18 runs
  • GitHub
  • Paper
  • License



Run time and cost

This model runs on Nvidia T4 GPU hardware.


DOVER: Disentangled Objective Video Quality Evaluator

WARNING: I could not replicate the precise results from the DOVER repo, despite copying their implementation directly and controlling random seeding.


DOVER is a specialized video quality assessment model for user-generated content, designed to analyze videos from both aesthetic and technical perspectives.


  • Dual Perspective Analysis: Separately evaluates aesthetic appeal and technical quality.
  • Pre-trained Model: Available on Hugging Face for easy integration.
  • High Accuracy: Utilizes advanced research to ensure precise quality assessments.

Use Case

Ideal for platforms hosting and managing user-generated video content, providing insights into both the technical and aesthetic quality of videos.