期刊
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
卷 33, 期 11, 页码 6494-6503出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TNNLS.2021.3081864
关键词
Feature extraction; Detectors; Convolution; Optimization; Object detection; Negative feedback; Semantics; iffDetector; inference-aware feature filtering (IFF); negative feedback; object detection
类别
资金
- Key Research and Development Program of Shandong Province [2019JZZY011101]
- National Natural Science Foundation of China [62076016, 61876015]
The study introduces a new feature optimization approach to enhance features and suppress background noise during both training and inference stages, utilizing a generic inference-aware feature filtering (IFF) module to achieve closed-loop feature optimization. Experimental results show that this method consistently outperforms state-of-the-art approaches on the PASCAL VOC and MS COCO datasets.
Modern convolutional neural network (CNN)-based object detectors focus on feature configuration during training but often ignore feature optimization during inference. In this article, we propose a new feature optimization approach to enhance features and suppress background noise in both the training and inference stages. We introduce a generic inference-aware feature filtering (IFF) module that can be easily combined with existing detectors, resulting in our iffDetector. Unlike conventional open-loop feature calculation approaches without feedback, the proposed IFF module performs the closed-loop feature optimization by leveraging high-level semantics to enhance the convolutional features. By applying the Fourier transform to analyze our detector, we prove that the IFF module acts as a negative feedback that can theoretically guarantee the stability of the feature learning. IFF can be fused with CNN-based object detectors in a plug-and-play manner with little computational cost overhead. Experiments on the PASCAL VOC and MS COCO datasets demonstrate that our iffDetector consistently outperforms state-of-the-art methods with significant margins.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据