Evaluating object detector ensembles for improving the robustness of artifact detection in endoscopic video streams
This work addresses artifact detection in endoscopic imaging for medical applications, but it is incremental as it combines existing models without introducing a fundamentally new approach.
The authors tackled the problem of detecting artifacts in endoscopic video streams by using an ensemble of two one-stage detectors (YOLOv4 and Yolact), which improved robustness while maintaining real-time computation, as shown by superior mean average precision over individual models and prior state-of-the-art methods.
In this contribution we use an ensemble deep-learning method for combining the prediction of two individual one-stage detectors (i.e., YOLOv4 and Yolact) with the aim to detect artefacts in endoscopic images. This ensemble strategy enabled us to improve the robustness of the individual models without harming their real-time computation capabilities. We demonstrated the effectiveness of our approach by training and testing the two individual models and various ensemble configurations on the "Endoscopic Artifact Detection Challenge" dataset. Extensive experiments show the superiority, in terms of mean average precision, of the ensemble approach over the individual models and previous works in the state of the art.