Skip to content

LiveFood

A food video dataset for highlight moment detection

Visit official site
Category AI models

What LiveFood is

LiveFood is described as a food video dataset for highlight moment detection, paired with a global prototype encoding model for incremental learning.

What you can do with it

  • Train a highlight moment detection model
  • Run domain incremental learning experiments
  • Use rich annotations for video understanding research

Pros & cons

✓ What we like

  • Annotations pass a strict double check
  • The dataset supports four distinct domains

! What to watch out for

  • Annotations were produced by professional workers
  • The described task centres on incremental learning

FAQ

How large is the dataset?

The material states more than 5100 food videos.

What is the GPE model for?

The material states it addresses the incremental learning problem.

Last reviewed: 2026-11-15

More LLM API platform tools

View all →

How we review