A visual dialogue between eye and machine

Can you imagine reading an image without touching it? This idea may seem impossible, but it has become a reality thanks to an algorithm that combines computer vision and natural language generation.

The model analyzes the image through CNN layers to extract visual features, then passes it through an RNN or Transformer network that generates an accurate descriptive sentence, and this sentence is converted into clear audio via a speech generation engine.

In one experiment, a blind child was used