A Mask-RCNN Baseline for Probabilistic Object Detection
This provides a baseline for understanding how standard detectors perform under a new evaluation metric, but it is incremental as it adapts an existing method.
The authors tackled the Probabilistic Object Detection Challenge by fine-tuning Mask-RCNN with post-processing, achieving second place with a score of 21.432 and the highest spatial and average quality.
The Probabilistic Object Detection Challenge evaluates object detection methods using a new evaluation measure, Probability-based Detection Quality (PDQ), on a new synthetic image dataset. We present our submission to the challenge, a fine-tuned version of Mask-RCNN with some additional post-processing. Our method, submitted under username pammirato, is currently second on the leaderboard with a score of 21.432, while also achieving the highest spatial quality and average overall quality of detections. We hope this method can provide some insight into how detectors designed for mean average precision (mAP) evaluation behave under PDQ, as well as a strong baseline for future work.