🚨 Baseline released, SiamMOT: Siamese Multi-Object Tracking

We are excited to share that the baseline for the competition is released!

The baseline is based on the SIAM-MOT model by AWS Rekognition and fetches EDR of 0.669 and AFDR of 0.6265! :tada:

The baseline is part of the starter kit and should help you tremendously with your submissions! :muscle:

Airborne Detection and Tracking Benchmark

Frame-level Airborne Object Detection Benchmark

You can start building on the top of the baseline and get a head start! :rocket:

:warning: Please note that identical SiamMOT models (with delta <= 1.5% in EDR or AFDR) will be disqualified from winning the prize. An identical model is a model that uses the exact same code and config file provided with the baseline.

1 Like

@shivam: any way to run the baseline in colab for local validation ?