search

NVlabs / FoundationPose

[CVPR 2024 Highlight] FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects
https://nvlabs.github.io/FoundationPose/
Other
1.3k stars 167 forks source link

I think there is something mixed directory usage on YCB-Video and LM dataset on `datareader.py.` #160

Open psh9002 opened 2 months ago

psh9002 commented 2 months ago

Hi

I am trying to run for the public dataset with YCB-Video. The run_nerf.py works well.

However, when I checked def run_pose_estimation() on run_ycb_video.py and then YcbVideoReader on datareader.py, it seems the defined variables for the directory are tangled.

The variables in the parent class BopBaseReader do not match the YCB-Video dataset structure on YcbVideoReader, neither the uploaded samples nor the official one.

If I misuse your code, please let me know the details for running the demo.

wenbowen123 commented 2 months ago

At which step did you run into errors?

QQH00 commented 2 months ago

I met the same issue, you should download the BOP format YCB-video dataset from here: https://bop.felk.cvut.cz/datasets/