Learning to Learn from Noisy Web Videos
arXiv (Cornell University)(2017)
摘要
Understanding the simultaneously very diverse and intricately fine-grained set of possible human actions is a critical open problem in computer vision. Manually labeling training videos is feasible for some action classes but doesn't scale to the full long-tailed distribution of actions. A promising way to address this is to leverage noisy data from web queries to learn new actions, using semi-supervised or "webly-supervised" approaches. However, these methods typically do not learn domain-specific knowledge, or rely on iterative hand-tuned data labeling policies. In this work, we instead propose a reinforcement learning-based formulation for selecting the right examples for training a classifier from noisy web search results. Our method uses Q-learning to learn a data labeling policy on a small labeled training dataset, and then uses this to automatically label noisy web data for new visual concepts. Experiments on the challenging Sports-1M action recognition benchmark as well as on additional fine-grained and newly emerging action classes demonstrate that our method is able to learn good labeling policies for noisy data and use this to learn accurate visual concept classifiers.
更多查看译文
关键词
data labeling policy,reinforcement learning,Q-learning,labeled training dataset,action classes,good labeling policies,computer vision,manually labeling training videos,long-tailed distribution,domain-specific knowledge,Web queries,action recognition benchmark,iterative hand-tuned data labeling policies,noisy Web videos
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络