📝 Image Captions
Overview
Each photo in FungiTastic includes a detailed, automatically generated caption describing visual features relevant for identification.
Details
- Generated by: Malmo-7B Vision-Language Model
- Description Focus: Color, shape, texture, size of the fungi and its parts.
- Prompt: Tailored to avoid speculation, highlight identification-relevant details.
Usage
- Provided as a field in the metadata for each image.
- Enables vision-language and multi-modal research.
- Useful for text-only classification and fusion models.
Example
"The image shows three mushrooms in a grassy area. They have white stems and light brown caps. Their caps are dome-shaped..."
Learn More
- Photographs for images linked to captions.
- VLM Fusion Models for downstream uses.