A react based web demo for Loomo Human-Robot-Interaction(HRI)
Top: Caption and VQA tasks based on the RGB-D video streams observed by Loomo.
Bottom: Visual Grounding task based on the reconstructed 3D scene.
LoomoHRI.mp4
Use npm to install the tool into your react project.
cd Loomo-Front-End
npm i Loomo-Front-End --save
Get SG results like 'localmode_locobot_output'. Run scripts in 'script' to get all related caption & VQA files.
- SG_convert: get annotation files & caption files.
- QA_generator: get QA files.
npm start
We are grateful to react-annotation-tool, on which our codes are developed.