DiagNet: Detecting Objects using Diagonal Constraints on Adjacency Matrix of Graph Neural Network
This addresses the problem of simplifying object detection pipelines for computer vision researchers by removing anchor box design, though it appears incremental as it builds on YOLO models.
The paper tackles object detection by proposing DiagNet, which uses diagonal constraints on adjacency matrices in graph neural networks to eliminate anchor boxes, achieving a 7.5% higher mAP50 on Pascal VOC than YOLOv1 and up to 5.1% higher mAP on MS COCO compared to YOLO variants.
We propose DaigNet, a new approach to object detection with which we can detect an object bounding box using diagonal constraints on adjacency matrix of a graph convolutional network (GCN). We propose two diagonalization algorithms based on hard and soft constraints on adjacency matrix and two loss functions using diagonal constraint and complementary constraint. The DaigNet eliminates the need for designing a set of anchor boxes commonly used. To prove feasibility of our novel detector, we adopt detection head in YOLO models. Experiments show that the DiagNet achieves 7.5% higher mAP50 on Pascal VOC than YOLOv1. The DiagNet also shows 5.1% higher mAP on MS COCO than YOLOv3u, 3.7% higher mAP than YOLOv5u, and 2.9% higher mAP than YOLOv8.