arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2007.06837cs.CVcs.LG
收藏

Few-Shot目标检测的Top-Related元学习方法

Top-Related Meta-Learning Method for Few-Shot Object Detection

Institute of Computing Technology(计算技术研究所);University of Chinese Academy of Sciences(中国科学院大学)

Qian Li, Nan Guo, Xiaochun Ye, Duo Wang, Dongrui Fan, Zhimin Tang

摘要:

提出了许多元学习方法用于Few-Shot检测。然而,以前的大多数方法都有两个主要问题,检测AP差,以及由于不平衡和数据集不足而导致的强偏差。以往的工作主要通过增加数据集、多关系注意力机制和子模块来缓解这些问题。然而,它们需要更多的成本。在这项工作中,对于元学习,我们发现主要的挑战集中在类别之间的相关或不相关的语义特征。因此,基于语义特征,我们提出了Top-C分类损失(即,TCL-C)的分类任务和基于类别的分组机制,基于类别的元特征获得的元模型。TCL-C利用真标签预测和最可能的C-1错误分类预测来提高对Few-Shot类的检测性能。根据相似的外观(即,视觉外观、形状和四肢等)和对象经常出现的环境,基于类别的分组机制将类别划分为不相交的组,使得组内类别之间相似的语义特征更加紧凑,组间差异更加显著,缓解了强偏见问题,进一步提高了检测AP。整个训练过程包括基本模型和微调阶段。在Pascal VOC数据集上的大量实验表明,将TCL-C与基于类别的分组相结合的方法在Few-Shot检测中的性能明显优于现有的方法。与以前的竞争基准相比,我们的检测AP提高了近4%的Few-Shot检测。

英文摘要:

Many meta-learning methods are proposed for few-shot detection. However, previous most methods have two main problems, poor detection APs, and strong bias because of imbalance and insufficient datasets. Previous works mainly alleviate these issues by additional datasets, multi-relation attention mechanisms and sub-modules. However, they require more cost. In this work, for meta-learning, we find that the main challenges focus on related or irrelevant semantic features between categories. Therefore, based on semantic features, we propose a Top-C classification loss (i.e., TCL-C) for classification task and a category-based grouping mechanism for category-based meta-features obtained by the meta-model. The TCL-C exploits the true-label prediction and the most likely C-1 false classification predictions to improve detection performance on few-shot classes. According to similar appearance (i.e., visual appearance, shape, and limbs etc.) and environment in which objects often appear, the category-based grouping mechanism splits categories into disjoint groups to make similar semantic features more compact between categories within a group and obtain more significant difference between groups, alleviating the strong bias problem and further improving detection APs. The whole training consists of the base model and the fine-tuning phases. According to grouping mechanism, we group the meta-features vectors obtained by meta-model, so that the distribution difference between groups is obvious, and the one within each group is less. Extensive experiments on Pascal VOC dataset demonstrate that ours which combines the TCL-C with category-based grouping significantly outperforms previous state-of-the-art methods for few-shot detection. Compared with previous competitive baseline, ours improves detection APs by almost 4% for few-shot detection.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑