AI Glossary

Visual grounding

What does “Visual grounding” mean?

Visual grounding: Connecting a textual phrase to a specific region or object in visual input.

Visual grounding visual guide: Connecting a textual phrase to a specific region or object in visual input.
Visual guide Open full size (new tab)

What is an example of “Visual grounding”?

An assistant identifies which of several people is described as wearing a red coat.

Keep learning

See also

Diffusion modelTermText-to-videoTermTemporal consistencyTermMotion transferTermImage captioningTerm

Same topic

Image maskTermReference image conditioningTermVoice activity detectionTermUpscalingTermVoice cloningTerm