Research Article
Boosting the Performance of Object Detection CNNs with Context-Based Anomaly Detection
@INPROCEEDINGS{10.1007/978-3-030-67537-0_11, author={Jan Blaha and George Broughton and Tom\^{a}š Krajn\^{\i}k}, title={Boosting the Performance of Object Detection CNNs with Context-Based Anomaly Detection}, proceedings={Collaborative Computing: Networking, Applications and Worksharing. 16th EAI International Conference, CollaborateCom 2020, Shanghai, China, October 16--18, 2020, Proceedings, Part I}, proceedings_a={COLLABORATECOM}, year={2021}, month={1}, keywords={Anomaly detection Object detection Context-aware neural networks Explainable neural networks Autoencoders}, doi={10.1007/978-3-030-67537-0_11} }- Jan Blaha
George Broughton
Tomáš Krajník
Year: 2021
Boosting the Performance of Object Detection CNNs with Context-Based Anomaly Detection
COLLABORATECOM
Springer
DOI: 10.1007/978-3-030-67537-0_11
Abstract
In this paper, we employ anomaly detection methods to enhance the ability of object detectors by using the context of their detections. This has numerous potential applications from boosting the performance of standard object detectors, to the preliminary validation of annotation quality, and even for robotic exploration and object search. We build our method on autoencoder networks for detecting anomalies, where we do not try to filter incoming data based on anomality score as is usual, but instead, we focus on the individual features of the data representing an actual scene. We show that one can teach autoencoders about the contextual relationship of objects in images, i.e. the likelihood of co-detecting classes in the same scene. This can then be used to identify detections that do and do not fit with the rest of the current observations in the scene. We show that the use of this information yields better results than using traditional thresholding when deciding if weaker detections are actually classed as observed or not. The experiments performed not only show that our method significantly improves the performance of CNN object detectors, but that it can be used as an efficient tool to discover incorrectly-annotated images.
