Out-of-Distribution Detection for Adaptive Computer Vision
This work addresses the issue of out-of-distribution detection for adaptive computer vision systems, which is incremental as it applies an existing method (normalizing flow) to a new application (camera parameter adaptation).
The paper tackles the problem of computer vision unreliability under unseen imaging conditions by proposing a method to adapt camera parameters using a normalizing flow-based out-of-distribution detector, resulting in an average increase of 3 to 4 percentage points in mAP, mAR, and F1 metrics for a YOLOv4 object detector in a small-scale study.
It is well known that computer vision can be unreliable when faced with previously unseen imaging conditions. This paper proposes a method to adapt camera parameters according to a normalizing flow-based out-of-distibution detector. A small-scale study is conducted which shows that adapting camera parameters according to this out-of-distibution detector leads to an average increase of 3 to 4 percentage points in mAP, mAR and F1 performance metrics of a YOLOv4 object detector. As a secondary result, this paper also shows that it is possible to train a normalizing flow model for out-of-distribution detection on the COCO dataset, which is larger and more diverse than most benchmarks for out-of-distibution detectors.