Detecting Owner-member Relationship with Graph Convolution Network in Fisheye Camera System

TL;DR

This paper introduces DeepWORD, a graph convolutional network-based method for detecting owner-member relationships in vehicle systems, overcoming occlusion challenges and demonstrating state-of-the-art accuracy and real-time performance.

Contribution

The paper presents DeepWORD, a novel GCN-based approach with a new dataset for owner-member relationship detection in fisheye camera systems, addressing occlusion and system applicability issues.

Findings

Achieved state-of-the-art accuracy in owner-member relationship detection.

Demonstrated real-time performance in embedded vehicle systems.

Provided a large-scale annotated dataset, WORD, for benchmarking.

Abstract

The owner-member relationship between wheels and vehicles contributes significantly to the 3D perception of vehicles, especially in embedded environments. However, to leverage this relationship we must face two major challenges: i) Traditional IoU-based heuristics have difficulty handling occluded traffic congestion scenarios. ii) The effectiveness and applicability of the solution in a vehicle-mounted system is difficult. To address these issues, we propose an innovative relationship prediction method, DeepWORD, by designing a graph convolutional network (GCN). Specifically, to improve the information richness, we use feature maps with local correlation as input to the nodes. Subsequently, we introduce a graph attention network (GAT) to dynamically correct the a priori estimation bias. Finally, we designed a dataset as a large-scale benchmark which has annotated owner-member relationship, called WORD. In the experiments we learned that the proposed method achieved state-of-the-art accuracy and real-time performance. The WORD dataset is made publicly available at https://github.com/NamespaceMain/ownermember-relationship-dataset.