Enhancing Egocentric Object Detection in Static Environments using Graph-based Spatial Anomaly Detection and Correction

TL;DR

This paper introduces a graph-based post-processing method that leverages spatial relationships in static environments to detect and correct object detection errors, significantly improving accuracy.

Contribution

It presents a novel graph neural network approach for anomaly detection and correction in egocentric object detection, utilizing spatial priors to enhance existing models.

Findings

Improved mAP@50 by up to 4% using the proposed method.

Effective correction of misclassifications and missed detections.

Applicable as a standalone or post-processing module for standard detectors.

Abstract

In many real-world applications involving static environments, the spatial layout of objects remains consistent across instances. However, state-of-the-art object detection models often fail to leverage this spatial prior, resulting in inconsistent predictions, missed detections, or misclassifications, particularly in cluttered or occluded scenes. In this work, we propose a graph-based post-processing pipeline that explicitly models the spatial relationships between objects to correct detection anomalies in egocentric frames. Using a graph neural network (GNN) trained on manually annotated data, our model identifies invalid object class labels and predicts corrected class labels based on their neighbourhood context. We evaluate our approach both as a standalone anomaly detection and correction framework and as a post-processing module for standard object detectors such as YOLOv7 and RT-DETR. Experiments demonstrate that incorporating this spatial reasoning significantly improves detection performance, with mAP@50 gains of up to 4%. This method highlights the potential of leveraging the environment's spatial structure to improve reliability in object detection systems.