Abstract
Despite recent advances in computer vision based on various convolutional architectures, video understanding remains an important challenge. In this work, we present and discuss a top solution for the large-scale video classification (labeling) problem introduced as a Kaggle competition based on the YouTube-8M dataset. We show and compare different approaches to preprocessing, data augmentation, model architectures, and model combination. Our final model is based on a large ensemble of video- and frame-level models but fits into rather limiting hardware constraints. We apply an approach based on knowledge distillation to deal with noisy labels in the original dataset and the recently developed mixup technique to improve the basic models.
Abstract (translated)
尽管最近基于各种卷积架构的计算机视觉方面取得了进展,但视频理解仍然是一项重要的挑战。在这项工作中,我们提出并讨论了基于YouTube-8M数据集作为Kaggle竞赛引入的大规模视频分类(标签)问题的顶级解决方案。我们展示并比较了预处理,数据增强,模型架构和模型组合的不同方法。我们的最终模型基于视频和帧级模型的大量集合,但适合于限制硬件约束。我们应用基于知识蒸馏的方法来处理原始数据集中的噪声标签和最近开发的混合技术以改进基本模型。
URL
https://arxiv.org/abs/1809.04403