Human recognition technology is a task that determines the people existing in images with the purpose of identifying them. However, automatic human recognition at night is still a challenge because of its need to align requirements with a high accuracy rate and speed. This article aims to design a novel approach that applies integrated face and gait analyses to enhance the performance of real-time human recognition in TIR images at night under various walking conditions. Therefore, a new network is proposed to improve the YOLOv3 model by fusing face and gait classifiers to identify individuals automatically. This network optimizes the TIR images, provides more accurate features (face, gait, and body segment) of the person, and possesses it through the PDM-Net to detect the person class; then, PRM-Net classifies the images for human recognition. The proposed methodology uses accurate features to form the face and gait signatures by applying the YOLO-face algorithm and YOLO algorithm. This approach was pre-trained on three night (DHU Night, FLIR, and KAIST) databases to simulate realistic conditions during the surveillance-protecting areas. The experimental results determined that the proposed method is superior to other results-related methods in the same night databases in accuracy and detection time.
Learning multi-scale feature representations is essential for dense prediction tasks in autonomous driving. Most existing works are based on U-shaped architectures, where high-resolution representations are progressively recovered by connecting different levels of the decoder with low-resolution representations from the encoder. We observed that rich details from low-level representation and high semantic information from high-level representations are not fully utilized in the cross-stage fusion process. Additionally, current architectures often struggle to extract efficient discriminative feature along object boundaries. To address this issue, we propose CSNet, a generic cross-stage subtraction network that extracts spatial and semantic multi-scale representations through guided contextual feature. This approach allows fine-grained features to refine deeper layers, capturing discriminative high-resolution features while filtering out redundant information. Specifically, we introduce a cross-stage subtraction module (CSM), which consists of three sub-modules: 1) a Short Path Unit, focusing on capturing complementary adjacent information; 2) Medium Path Unit for effective middle-stages features aggregation; and 3) Long Path Unit for redundant information masking and long-range context modeling. Additionally, we propose the Semantic Guided Context Reasoning (SGCR) module to reason and model contextual relations between different subtraction units. CSNet demonstrates consistent performance gains across various semantic segmentation datasets. Our model, CSNet-M, achieves 82.2% mIoU on the Camvid dataset, while CSNet-S and CSNet-M attain 79.6% and 80.5% mIoU accuracy, respectively, on the Cityscapes dataset. These results show that the proposed CSNet has the potential for enhancing real-time semantic segmentation in autonomous driving applications, offering improved accuracy and efficiency in diverse urban scenarios. The source code for this work will be published at https://github.com/mohamedac29/CSNet.