Google Introduces Promptable Gaze Target Estimation with Text and Visual Prompts

google · hf · 2026-08-14

Google released Gaze Target Estimation Anywhere with Concepts on Hugging Face.

This new promptable paradigm integrates subject localization and gaze estimation into an end-to-end transformer model. Without relying on multi-stage pipelines, the model uses text or visual prompts to identify subjects and predict their gaze targets directly.

Original post →

More from Research

Research channel →