Semantic Information for Object Detection

TL;DR

This paper introduces a knowledge-aware re-optimization approach for object detection in traffic scenes, utilizing a novel knowledge graph and semantic consistency to improve detection accuracy.

Contribution

It presents a new method for extracting and integrating a knowledge graph with semantic consistency models for object detection.

Findings

Limited but consistent improvements in precision and recall

Effective across different models like Faster-RCNN and DETR

Demonstrates the value of semantic knowledge integration in complex scenes

Abstract

In this paper, we demonstrate that the concept of Semantic Consistency and the ensuing method of Knowledge-Aware Re-Optimization can be adapted for the problem of object detection in intricate traffic scenes. Furthermore, we introduce a novel method for extracting a knowledge graph from a dataset of images provided with instance-level annotations, and integrate this new knowledge graph with the existing semantic consistency model. Combining both this novel hybrid knowledge graph and the preexisting methods of frequency analysis and external knowledge graph as sources for semantic information, we investigate the effectiveness of knowledge-aware re-optimization on the Faster-RCNN and DETR object detection models. We find that limited but consistent improvements in precision and or recall can be achieved using this method for all combinations of model and method studied.