Computer vision · Accessibility

More expressive signing avatars

Connecting facial motion capture, animation, and the non-manual elements of sign language.

Doctoral research figure showing a realistic signing avatar, head views and facial mesh
Model views from the doctoral work on facial animation for signing avatars.

Facial motion carries meaning

In sign languages, communication extends beyond hand movements. Facial expressions and other non-manual elements can carry linguistic information. A signing avatar needs to account for these elements as part of the communication itself.

From motion to animation

My doctoral work at UNICAMP investigated modeling and synthesizing facial aspects of Brazilian Sign Language for signing avatars. Related work explored facial motion capture, marker placement, and mapping captured movements to a realistic avatar.

Facial animation rig and controller positions used in the signing-avatar research
Facial controls support the synthesis of expressive avatar movements.

Read the doctoral thesis ↗

A connected research thread

The work brings together computer vision, computer graphics, language, and accessibility. It continues to inform the human-centered questions I bring to applied AI.

Explore related publications →
LET’S CONNECT

A useful project starts with a clear question.

Explore related research, practical learning, or an idea we could work on together.

Get in touch

Find your next idea.

Search across courses, research, projects, and resources.

Ideas, connected.esc to close