AI engineering glossary · Generative media multimodal
What is visual grounding?
Visual grounding: Linking words or phrases in text to specific regions in an image, usually by outputting bounding boxes or masks. It lets a model point at 'the person in the blue kurta' instead of just describing the scene.
Visual grounding: an example
Asked to find 'the red bicycle', the model returns box coordinates around that bicycle.
Related terms
More Generative media multimodal terms
- Visual instruction tuning
- Visual question answering
- VITS
- Voice activity detection
- Voice cloning
- VQ-VAE
- Wasserstein GAN
- Watermarking generated content
Where you learn it
Visual grounding is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.