通过转移分类权值的尺度化目标检测
Scaling Object Detection by Transferring Classification Weights
摘要：
大规模的目标检测数据集在类别和注释计数方面不断增加其大小。然而，在检测数据集中标注的对象级类别的数量比图像级分类标签小一个数量级。最先进的目标检测模型是以监督的方式训练的，这限制了它们可以检测的对象类别的数量。在本文中，我们提出了一种新的权重转移网络（WTN），可以有效地将知识从分类网络的权重转移到检测网络的权重，从而在没有框监督的情况下检测新的类。我们首先引入输入和特征规范化方案，以抑制vanilla WTN训练过程中的欠拟合。然后，我们提出了自动编码器-WTN（AE-WTN），它使用重建损失来保留目标潜在空间中所有类别的分类网络信息，以确保对新类别的泛化。与vanilla WTN相比，AE-WTN在两个Open Images评估集上获得了6%的绝对性能增益，分别为500个可见类和57个新类，在Visual Genome评估集上获得了25%的绝对性能增益，具有200个新类。该代码可在https://github.com/xternalz/AE-WTN上获得。
英文摘要：
Large scale object detection datasets are constantly increasing their size in terms of the number of classes and annotations count. Yet, the number of object-level categories annotated in detection datasets is an order of magnitude smaller than image-level classification labels. State-of-the art object detection models are trained in a supervised fashion and this limits the number of object classes they can detect. In this paper, we propose a novel weight transfer network (WTN) to effectively and efficiently transfer knowledge from classification network's weights to detection network's weights to allow detection of novel classes without box supervision. We first introduce input and feature normalization schemes to curb the under-fitting during training of a vanilla WTN. We then propose autoencoder-WTN (AE-WTN) which uses reconstruction loss to preserve classification network's information over all classes in the target latent space to ensure generalization to novel classes. Compared to vanilla WTN, AE-WTN obtains absolute performance gains of 6% on two Open Images evaluation sets with 500 seen and 57 novel classes respectively, and 25% on a Visual Genome evaluation set with 200 novel classes. The code is available at https://github.com/xternalz/AE-WTN.
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。
