论文标题
Omni源自Weybly的监督学习，以供视频识别
Omni-sourced Webly-supervised Learning for Video Recognition
论文作者
论文摘要
我们介绍了Omnisource，这是一个新颖的框架，用于利用Web数据训练视频识别模型。 Omnisource克服了数据格式之间的障碍，例如图像，简短视频和长期未经修剪的视频，以供您进行韦伯利的学习。首先，具有多种格式的数据样本，通过特定于任务的数据收集策划并通过教师模型自动过滤，被转换为统一形式。然后提出了一种联合培训策略来处理Webly Sypersed学习中多个数据源和格式之间的域差距。联合培训中采用了几种良好的做法，包括数据平衡，重新采样和跨数据库混合。实验表明，通过利用来自多种来源和格式的数据，Omnisource在培训方面具有更高的数据效率。我们的模型只有350万张图像和80万分钟的视频从互联网上爬行（不到先前作品的2％），因此在Omnisource上学习了2D和3D-CONVNET基线模型的TOP-1准确性，分别在Kinetics-400 Benchmark上提高了3.0％和3.9％。借助Omnisource，我们建立了新的记录，并具有不同的视频识别策略。我们的最佳模型在动力学400基准上分别实现了80.4％，80.5％和83.6 TOP-1精确度，用于训练，用于训练，Imagenet预训练和IG-65M预培训。
We introduce OmniSource, a novel framework for leveraging web data to train video recognition models. OmniSource overcomes the barriers between data formats, such as images, short videos, and long untrimmed videos for webly-supervised learning. First, data samples with multiple formats, curated by task-specific data collection and automatically filtered by a teacher model, are transformed into a unified form. Then a joint-training strategy is proposed to deal with the domain gaps between multiple data sources and formats in webly-supervised learning. Several good practices, including data balancing, resampling, and cross-dataset mixup are adopted in joint training. Experiments show that by utilizing data from multiple sources and formats, OmniSource is more data-efficient in training. With only 3.5M images and 800K minutes videos crawled from the internet without human labeling (less than 2% of prior works), our models learned with OmniSource improve Top-1 accuracy of 2D- and 3D-ConvNet baseline models by 3.0% and 3.9%, respectively, on the Kinetics-400 benchmark. With OmniSource, we establish new records with different pretraining strategies for video recognition. Our best models achieve 80.4%, 80.5%, and 83.6 Top-1 accuracies on the Kinetics-400 benchmark respectively for training-from-scratch, ImageNet pre-training and IG-65M pre-training.
