Uncertainty for Identifying Open-Set Errors in Visual Object Detection

Deployed into an open world, object detectors are prone to open-set errors, false positive detections of classes not present in the training dataset.We propose GMM-Det, a real-time method for extracting epistemic uncertainty from identify and reject errors. GMM-Det trains detector produce structured logit space that is modelled with class-specific Gaussian Mixture Models. At test time, errors identified by their low log-probability under all We two common architectures, Faster R-CNN RetinaNet, across three varied datasets spanning robotics computer vision. Our results show consistently outperforms existing techniques identifying rejecting detections, especially at low-error-rate operating point required safety-critical applications. maintains detection performance, introduces only minimal computational overhead. also introduce methodology converting specific open-set evaluate performance detection.