image-text model

noun

UK /ˈɪmɪdʒ tɛkst ˈmɒdl/

US /ˈɪmɪdʒ tɛkst ˈmɑːdl/

A computer program that learns by looking at pictures and reading words together.

The image-text model can tell what is happening in a photo by reading the caption next to it.

SYNONYMS

vision-language model multimodal model image-caption model visual-language model

RELATED WORDS

artificial intelligence machine learning neural network computer vision natural language processing

OTHER MEANINGS

2. A system that matches pictures with the words that describe them.

The image-text model found the right caption for each picture in the test.

3. A tool that can answer questions about a photo using written language.

She asked the image-text model what color the car in the picture was.

4. A kind of AI that understands both images and text at the same time.

Researchers trained the image-text model on millions of photo-and-sentence pairs.