hYOLO Model: Enhancing Object Classification with Hierarchical Context in YOLOv8

TL;DR

This paper introduces hYOLO, a hierarchical object detection and classification model built on YOLOv8, which leverages object relationships and hierarchy to improve accuracy and contextual understanding in images.

Contribution

The paper proposes a novel hierarchical architecture, a modified loss function, and a new performance metric for YOLO-based models to better capture object relationships and hierarchy.

Findings

Improved classification accuracy on hierarchical datasets

Effective modeling of object relationships in detection tasks

Enhanced contextual understanding over flat classification methods

Abstract

Current convolution neural network (CNN) classification methods are predominantly focused on flat classification which aims solely to identify a specified object within an image. However, real-world objects often possess a natural hierarchical organization that can significantly help classification tasks. Capturing the presence of relations between objects enables better contextual understanding as well as control over the severity of mistakes. Considering these aspects, this paper proposes an end-to-end hierarchical model for image detection and classification built upon the YOLO model family. A novel hierarchical architecture, a modified loss function, and a performance metric tailored to the hierarchical nature of the model are introduced. The proposed model is trained and evaluated on two different hierarchical categorizations of the same dataset: a systematic categorization that disregards visual similarities between objects and a categorization accounting for common visual characteristics across classes. The results illustrate how the suggested methodology addresses the inherent hierarchical structure present in real-world objects, which conventional flat classification algorithms often overlook.