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
To test the perception part, we can use our gimballed camera on the ground to take videos of the test quad flying, and we can sync up the ground truth trajectory to the video frames based on timestamps.
The text was updated successfully, but these errors were encountered:
Something like this, where we do object detection to just get bounding boxes of other drones: https://arxiv.org/pdf/2305.09972.pdf
What they did in the paper was pretrain on this dataset: https://universe.roboflow.com/new-workspace-0k81p/flying_object_dataset
Then transfer-learn on this dataset: https://universe.roboflow.com/ahmedmohsen/drone-detection-new-peksv
To test the perception part, we can use our gimballed camera on the ground to take videos of the test quad flying, and we can sync up the ground truth trajectory to the video frames based on timestamps.
The text was updated successfully, but these errors were encountered: