Skip to content

Dataset · myned-ai

FLAME and GNM 3D head meshes for Ava-256 dataset

myned-ai/Ava256-FLAME-GNM

Parameters and 3D head meshes including teeth and tongue for 15,345 multi-view frames of 256 individuals.

What's inside

The dataset contains 3D head reconstructions for 15,345 frames from the Meta Ava-256 subset, featuring 256 subjects filmed across 80 cameras at roughly 60 frames per person. Provided in a tabular-text format, each frame includes meshes and parameters for FLAME and Google GNM models complete with teeth, gums, and tongue, alongside error metrics relative to lip and teeth landmarks. Raw photographs are not included.

What you can do with it

  • Training 3D facial reconstruction and animation models with detailed oral cavity rendering
  • Comparing accuracy between FLAME and GNM parametric head models
  • Creating digital avatars and facial tracking systems
Size
10K–100K rows
Downloads this month
1.2K
Likes
16

How to load it

from datasets import load_dataset

ds = load_dataset("myned-ai/Ava256-FLAME-GNM")