当深度学习遇到大视频数据

[摘要] 视频信号是大数据中的大数据，这种海量视频数据带来了存储、传输、处理、管理等方面的挑战，同时也提供了大量有价值的信息和商业机会。认为深度学习颠覆了视觉理解的进程，从图像分类到物体检测、语义分割等更细更复杂的任务，从视频里物体的检测与跟踪到物体属性和行为的分析，特别是关于人和车的理解技术。指出随着计算能力和大数据持续快速增长，加上深度学习、主动学习、迁移学习、无监督学习、强化学习等强大机器学习技术继续发展，让机器可以像人一样看到并理解世界的前景是乐观的。

[关键词] 深度学习；大视频数据；人工智能

[Abstract] As the biggest big data, big video data imposes significant challenges to storage, transmission, processing, and management, but it also provides an enormous amount of rich information and huge business opportunities. Deep learning has fundamentally changed the way we understand visual information, from image classification to object detection and semantic segmentation, and from object detection and tracking to the analysis of object attributes and behaviors in video. This is especially true in understanding humans and vehicles. With the rapid growth of big compute, big data, and advanced machine learning techniques such as deep learning, active learning, transfer learning, unsupervised learning, and reinforcement learning, machines will soon be able to see and understand the physical world in the same way as humans do.

[Keywords] deep learning; big video data; artificial intelligence