YouTube-8M A Large-Scale Video Classification Benchmark英文文献资料.pdfVIP

YouTube-8M: A Large-Scale Video Classification Benchmark Sami Abu-El-Haija Nisarg Kothari Joonseok Lee Paul Natsev haija@ ndk@ joonseok@ natsev@ George Toderici Balakrishnan Varadarajan Sudheendra Vijayanarasimhan gtoderici@ balakrishnanv@ svnaras@ Google Research ABSTRACT Many recent advancements in Computer Vision are attributed to large datasets. Open-source software packages for Machine Learn- ing and inexpensive commodity hardware have reduced the bar- rier of entry for exploring novel approaches at scale. It is possible to train models over millions of examples within a few days. Al- though large-scale datasets exist for image understanding, such as ImageNet, there are no comparable size video classification datasets. In this paper, we introduce YouTube-8M, the largest multi-label video classification dataset, composed of million videos— hours of video—annotated with a vocabulary of visual en- tities. To get the videos and their (multiple) labels, we used a YouTube video annotation system, which labels videos with the main topics in them. While the labels are machine-generated, they have high-precision and are derived from a variety of human-based signals including metadata and query click signals, so they repre- sent an excellent target for content-based annotation approaches. Figure 1: YouTube-8M is a large-scale benchmark for general We filtered the video labels (Knowledge Graph entities) using both