Lightweight Feature Enhancement Network for Single-Shot Object Detection

Abstract

At present, the one-stage detector based on the lightweight model can achieve real-time speed, but the detection performance is challenging. To enhance the discriminability and robustness of the model extraction features and improve the detector’s detection performance for small objects, we propose two modules in this work. First, we propose a receptive field enhancement method, referred to as adaptive receptive field fusion (ARFF). It enhances the model’s feature representation ability by adaptively learning the fusion weights of different receptive field branches in the receptive field module. Then, we propose an enhanced up-sampling (EU) module to reduce the information loss caused by up-sampling on the feature map. Finally, we assemble ARFF and EU modules on top of YOLO v3 to build a real-time, high-precision and lightweight object detection system referred to as the ARFF-EU network. We achieve a state-of-the-art speed and accuracy trade-off on both the Pascal VOC and MS COCO data sets, reporting 83.6% AP at 37.5 FPS and 42.5% AP at 33.7 FPS, respectively. The experimental results show that our proposed ARFF and EU modules improve the detection performance of the ARFF-EU network and achieve the development of advanced, very deep detectors while maintaining real-time speed.

Lightweight Saliency Target Intelligent Detection based on Multi-scale Feature Adaptive Fusion

“…Currently, in-depth research-based target detection technology has been applied to autonomous driving systems; face recognition; medical images; security; fault diagnosis; military and other field [3]. The object detection method identifies the most attractive targets from the input image and is the initial step of a multi-vision computer [4]. When considering the evolution of saliency target detection, it can be categorized into two distinct approaches: traditional methods that rely on manually crafted features and heuristic priors, and task-oriented saliency target detection methods built upon deep learning.…”