Attention-Enhanced YOLOv8 for Accurate Pedestrian Detection and Count Estimation

Autonomous vehicles (AVs) are crucial for improving the safety of highways by reducing man-made errors that account for the most dominant source of road accidents.They also offer significant potential for increased efficiency in transportation, lowering emissions, and enhancing mobility for the elderly and disabled.However, significant obstacles still need more study to be overcome.For example, in crowded metropolitan settings, AVs must correctly sense their surroundings to operate safely.Numerous research papers explore effective methods for precisely determining the surroundings of AVs.However, there are still challenges to detecting non-static objects, especially pedestrians.In this article, deep learning algorithms have been employed to realize the real-time detection of pedestrians, resulting in the advancement of AVS development.By enhancing feature extraction within the detection algorithm, This paper presents the Convolutional Block Attention Module (CBAM-YOLOv8) model, which integrates three CBAM modules into the backbone network of YOLOv8.Additionally, the DUA-YOLOv8 and ECA-YOLOv8 models will be introduced.The results obtained from the experiments on a combined dataset of 1520 images collected from the INRIA and ETH datasets indicate that CBAM-YOLOv8 slightly improves recall and [email protected]:0.95.The proposed model reached a [email protected]:0.95 of 0.57 and 42 FPS on an NVIDIA Tesla T4 GPU.The evaluation metrics for CBAM-YOLOv8 showed greater enhancements compared to SE-YOLOv8 and ECA-YOLOv8.