arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.17115cs.CV
收藏

基于上下文感知聚合的Few-Shot目标检测稠密关系提取

Dense Relation Distillation with Context-aware Aggregation for Few-Shot Object Detection

Peking University(北京大学);Beijing University of Posts and Telecommunications(北京邮电大学)

Hanzhe Hu, Shuai Bai, Aoxue Li, Jinshi Cui, Liwei Wang

摘要:

用于目标检测的传统的基于深度学习的方法需要大量的限界框注释用于训练,这对于获得这样的高质量注释数据是昂贵的。Few-Shot目标检测是一种非常具有挑战性的方法,它只需要很少的注释样本就可以适应新的类,因为新对象的细粒度特征很容易被忽略,只有很少的数据可用。在这项工作中,为了充分利用带注释的新对象的特征并捕获查询对象的细粒度特征,我们提出了具有上下文感知聚合的密集关系蒸馏(DCNet)来解决少镜头检测问题(Few-Shot detection problem)。基于元学习的框架,稠密关系蒸馏模块的目标是充分利用支持特征,其中支持特征和查询特征是密集匹配的,以前馈方式覆盖所有空间位置。引导信息的大量使用赋予模型处理常见挑战的能力,例如外观变化和遮挡。此外,为了更好地捕获尺度感知特征,上下文感知聚合模块自适应地利用不同尺度的特征,以获得更全面的特征表示。大量的实验表明,我们提出的方法在PASCAL VOC和MS COCO数据集上实现了最先进的结果。代码将在https://github.com/hzhupku/DCNet上提供。

英文摘要:

Conventional deep learning based methods for object detection require a large amount of bounding box annotations for training, which is expensive to obtain such high quality annotated data. Few-shot object detection, which learns to adapt to novel classes with only a few annotated examples, is very challenging since the fine-grained feature of novel object can be easily overlooked with only a few data available. In this work, aiming to fully exploit features of annotated novel object and capture fine-grained features of query object, we propose Dense Relation Distillation with Context-aware Aggregation (DCNet) to tackle the few-shot detection problem. Built on the meta-learning based framework, Dense Relation Distillation module targets at fully exploiting support features, where support features and query feature are densely matched, covering all spatial locations in a feed-forward fashion. The abundant usage of the guidance information endows model the capability to handle common challenges such as appearance changes and occlusions. Moreover, to better capture scale-aware features, Context-aware Aggregation module adaptively harnesses features from different scales for a more comprehensive feature representation. Extensive experiments illustrate that our proposed approach achieves state-of-the-art results on PASCAL VOC and MS COCO datasets. Code will be made available at https://github.com/hzhupku/DCNet.

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。

↑