论文标题
端到端的搜索在聚合数据集上依次培训
End-to-end Person Search Sequentially Trained on Aggregated Dataset
论文作者
论文摘要
在视频监视应用程序中，人员搜索是一项具有挑战性的任务，包括检测人员和从轮廓中提取功能以重新识别（RE-ID）目的。我们提出了一个新的端到端模型，该模型可以通过单个深层卷积神经网络体系结构共同计算检测和提取步骤。共享两个任务之间的特征图，以共同描述人们的共同点和特殊性，可以更快地运行时，这在现实世界应用程序中很有价值。除了达到最先进的精度外，该多任务模型还可以是逐个训练的任务，从而更广泛接受输入数据集类型。确实，我们表明，在没有昂贵的身份注释的情况下汇总更多的行人检测数据集会使共享特征地图更加通用，并提高了重新确定的精度。此外，这些增强的共享特征映射导致重新ID特征更强大，使得跨数据库方案。
In video surveillance applications, person search is a challenging task consisting in detecting people and extracting features from their silhouette for re-identification (re-ID) purpose. We propose a new end-to-end model that jointly computes detection and feature extraction steps through a single deep Convolutional Neural Network architecture. Sharing feature maps between the two tasks for jointly describing people commonalities and specificities allows faster runtime, which is valuable in real-world applications. In addition to reaching state-of-the-art accuracy, this multi-task model can be sequentially trained task-by-task, which results in a broader acceptance of input dataset types. Indeed, we show that aggregating more pedestrian detection datasets without costly identity annotations makes the shared feature maps more generic, and improves re-ID precision. Moreover, these boosted shared feature maps result in re-ID features more robust to a cross-dataset scenario.
