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 1906.03019cs.CVcs.LG
收藏

通过多任务和多数据集学习的视觉人理解

Visual Person Understanding through Multi-Task and Multi-Dataset Learning

RWTH Aachen University(亚琛工业大学);Visual Computing Institute, RWTH Aachen University(亚琛工业大学视觉计算研究所)

Kilian Pfeiffer, Alexander Hermans, István Sárándi, Mark Weber, Bastian Leibe

摘要:

我们解决的问题,学习一个单一的模型的人重新识别,属性分类,身体部位分割,姿势估计。通过对这些任务的预测,我们可以更全面地了解人,这对许多应用程序都很有价值。这是一个典型的多任务学习问题。然而,不存在可以从这些任务中联合学习的数据集。因此,在训练过程中需要组合多个数据集,这在其他情况下通常会导致过去的性能下降。我们广泛地评估了不同的任务和数据集如何相互影响,以及任务之间不同程度的参数共享如何影响性能。我们的最终模型匹配或优于其单任务同行,而不会产生显着的计算开销,使其非常有趣的资源受限的情况下,如移动机器人。

英文摘要:

We address the problem of learning a single model for person re-identification, attribute classification, body part segmentation, and pose estimation. With predictions for these tasks we gain a more holistic understanding of persons, which is valuable for many applications. This is a classical multi-task learning problem. However, no dataset exists that these tasks could be jointly learned from. Hence several datasets need to be combined during training, which in other contexts has often led to reduced performance in the past. We extensively evaluate how the different task and datasets influence each other and how different degrees of parameter sharing between the tasks affect performance. Our final model matches or outperforms its single-task counterparts without creating significant computational overhead, rendering it highly interesting for resource-constrained scenarios such as mobile robotics.

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。

↑