BundleTrack icon indicating copy to clipboard operation
BundleTrack copied to clipboard

Inference on custom dataset

Open guillermodecelisrodriguez opened this issue 2 years ago • 0 comments

First of all thank you for sharing your incredible work!!

I am trying to perform inference on data from YCBInEOAT ("006_mustard_bottle") and the script run_ycbineoat.py seems to work but this is the output from the command line:

data has 737 images Connected to port 5555 color file: /home//Desktop/BundleTrack/YCBInEOAT/mustard0/rgb/1581120424100262102.png

New frame 1581120424100262102 zmq start waiting for reply zmq got reply color file: /home//Desktop/BundleTrack/YCBInEOAT/mustard0/rgb/1581120424148532296.png

New frame 1581120424148532296 zmq start waiting for reply zmq got reply finding corres between 1581120424148532296(id=1) and 1581120424100262102(id=0) frame 1581120424148532296 and 1581120424100262102 findNN too few match, 1581120424148532296 status marked as FAILcolor file: /home//Desktop/BundleTrack/YCBInEOAT/mustard0/rgb/1581120424213125141.png

New frame 1581120424213125141 zmq start waiting for reply zmq got reply finding corres between 1581120424213125141(id=1) and 1581120424100262102(id=0) frame 1581120424213125141 and 1581120424100262102 findNN too few match, 1581120424213125141 status marked as FAILcolor file: /home//Desktop/BundleTrack/YCBInEOAT/mustard0/rgb/1581120424269362478.png

And continues like this for all the images located in the directory without saving any results on the corresponding results folder.

I would appreciate any help, thanks in advance!!