Detecting Reflections by Combining Semantic and Instance Segmentation
This addresses accuracy issues in computer vision tasks like detection and segmentation for surveillance applications, but is an incremental application of existing panoptic segmentation methods.
The paper tackled the problem of reflection-induced false positives in automated detection systems by fusing instance and semantic segmentation to identify reflections without explicit labeling. The approach reduced false positive detections in real-world surveillance data with many reflective surfaces.
Reflections in natural images commonly cause false positives in automated detection systems. These false positives can lead to significant impairment of accuracy in the tasks of detection, counting and segmentation. Here, inspired by the recent panoptic approach to segmentation, we show how fusing instance and semantic segmentation can automatically identify reflection false positives, without explicitly needing to have the reflective regions labelled. We explore in detail how state of the art two-stage detectors suffer a loss of broader contextual features, and hence are unable to learn to ignore these reflections. We then present an approach to fuse instance and semantic segmentations for this application, and subsequently show how this reduces false positive detections in a real world surveillance data with a large number of reflective surfaces. This demonstrates how panoptic segmentation and related work, despite being in its infancy, can already be useful in real world computer vision problems.