FireRescue: UAV-Based Object Detection for Fire Rescue

Published:Dec 31, 2025 04:37
1 min read
ArXiv

Analysis

This paper addresses a critical gap in fire rescue research by focusing on urban rescue scenarios and expanding the scope of object detection classes. The creation of the FireRescue dataset and the development of the FRS-YOLO model are significant contributions, particularly the attention module and dynamic feature sampler designed to handle complex and challenging environments. The paper's focus on practical application and improved detection performance is valuable.

Reference

The paper introduces a new dataset named "FireRescue" and proposes an improved model named FRS-YOLO.