Joas Pambou's new application aims to enhance conversational AI by building a model that not only describes images but engages users in interactive discussions.
Previously, we developed an application that provided audio descriptions of images for visually impaired users by combining image-to-text and text-to-speech models.
The focus of this next application is to take AI interaction further by allowing users to ask detailed questions and learn more about their visual content.
We'll utilize the LLaVA model, which integrates both image understanding and conversational capabilities, turning our tool into a comprehensive multimodal interactive assistant.
[
Collection
]
[
|
...
]