search

nuoxu / AKGVP

Aligning Knowledge Graph with Visual Perception for Object-goal Navigation (ICRA 2024)
MIT License
24 stars 3 forks source link

Clip images as input #3

Closed rginjapan closed 6 months ago

rginjapan commented 6 months ago

Sorry, I think the code is not consistency with your paper, since the clip_scene has not been used in your code... https://github.com/nuoxu/AKGVP/blob/19c9fb751f5cf9154557365a990ba790159a539b/models/AKGVP.py#L123C9-L123C71