LSTD: A Low-Shot Transfer Detector for Object Detection

Authors

  • Hao Chen Huazhong University of Science and Technology
  • Yali Wang Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences
  • Guoyou Wang Huazhong University of Science and Technology
  • Yu Qiao Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences

Keywords:

Object Detection, Low-Shot Learning, Transfer Learning

Abstract

Recent advances in object detection are mainly driven by deep learning with large-scale detection benchmarks. However, the fully-annotated training set is often limited for a target detection task, which may deteriorate the performance of deep detectors. To address this challenge, we propose a novel low-shot transfer detector (LSTD) in this paper, where we leverage rich source-domain knowledge to construct an effective target-domain detector with very few training examples. The main contributions are described as follows. First, we design a flexible deep architecture of LSTD to alleviate transfer difficulties in low-shot detection. This architecture can integrate the advantages of both SSD and Faster RCNN in a unified deep framework. Second, we introduce a novel regularized transfer learning framework for low-shot detection, where the transfer knowledge (TK) and background depression (BD) regularizations are proposed to leverage object knowledge respectively from source and target domains, in order to further enhance fine-tuning with a few target images. Finally, we examine our LSTD on a number of challenging low-shot detection experiments, where LSTD outperforms other state-of-the-art approaches. The results demonstrate that LSTD is a preferable deep detector for low-shot scenarios.

Downloads

Published

2018-04-29

How to Cite

Chen, H., Wang, Y., Wang, G., & Qiao, Y. (2018). LSTD: A Low-Shot Transfer Detector for Object Detection. Proceedings of the AAAI Conference on Artificial Intelligence, 32(1). Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/11716