You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm looking for ways to contribute to the autoware perception stack by improving the accuracy of the models. Are the training datasets available for any of these modules?
I did some ad hoc testing against the traffic_light_classifier model with GPU/CNN enabled. It worked fairly well on a small batch of real-world image data that I collected using my mobile device, but it gave incorrect predictions on some images I downloaded from google images. I could provide the test samples if they'd be useful.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
I'm looking for ways to contribute to the autoware perception stack by improving the accuracy of the models. Are the training datasets available for any of these modules?
I did some ad hoc testing against the traffic_light_classifier model with GPU/CNN enabled. It worked fairly well on a small batch of real-world image data that I collected using my mobile device, but it gave incorrect predictions on some images I downloaded from google images. I could provide the test samples if they'd be useful.
Was tensorrt_yolo trained on the coco dataset or a custom dataset?
Beta Was this translation helpful? Give feedback.
All reactions