Skip to content

Dataset · Fsoft-AIC

Multimodal Robot Design Dataset

Fsoft-AIC/RobotDesign1M

Millions of image-text pairs from research papers for robot design generation and analysis.

What's inside

A large multimodal corpus with a volume ranging from 1 to 10 million records, gathered from scientific literature across various fields of robotics. It contains images and text descriptions intended for training base models in engineering design comprehension.

What you can do with it

  • Training generative text-to-image models to create robot concepts and blueprints
  • Training multimodal Visual Question Answering (VQA) models to answer questions on robot designs
  • Creating image retrieval systems for robotic components using text queries
Tasks
Image text → text, Text → image
Size
1M–10M rows
Downloads this month
72.7K
Likes
16

How to load it

from datasets import load_dataset

ds = load_dataset("Fsoft-AIC/RobotDesign1M")