An Enhanced Light Object Detection for Indiscernible Object in the Special Scene

  • Quanyou Zhang Chongqing University, Chongqing, China
  • Yong Feng Chongqing University, Chongqing, China
  • Yong-heng Wang Zhejiang Lab, Yuhang district, Hangzhou, China
  • Bao-hua Qiang Guangxi Key Laboratory of Trusted Software, Guilin University of Electronic Technology, Guilin, China
  • Lufeng Wang Chongqing Industry Polytechnic College, Chongqing, China
  • Zebin Zhang Zhengzhou university, Zhengzhou City, China


We present a transfer learning method named Special Application Transfer (SAT) for special object detection in a real life scenario. Our method improves fine-tuning hyper-parameter and adds unrecognized samples to detect special samples when training object detection neural networks for classification. We implement the model of NanoDet on special supervised datasets and fine-tune the hyper-parameter on a target task. More importantly, we combine a few carefully selected samples in training and simple heuristic fine-tuning to achieve good performance on special object detection in real-life scenarios. Our method (SAT) performs well across surprisingly the small dataset the medium dataset and the large dataset. SAT achieves 95% AP (Average Precision) on the small dataset, 94.8% AP on the medium dataset, and 94.5% AP on the large dataset. The performances of AP run-time and training convergence are perfect, compared with the original method and well-established famous methods on the challenging COCO dataset and our dataset. We hope our work could promote and complete the practical application in more real life scenarios. Our code is available at: