arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2012.13726cs.CV
收藏

直接从频域快速准确地识别压缩视频动作

Faster and Accurate Compressed Video Action Recognition Straight from the Frequency Domain

Samuel Felipe dos Santos, Jurandy Almeida

摘要:

人类动作识别由于其广泛的应用,如监控、医疗、工业环境、灵巧家庭等,已经成为计算机视觉中最活跃的研究领域之一。最近,深度学习已成功用于学习强大且可解释的特征,以识别视频中的人类动作。大多数现有的深度学习方法已经被设计用于将视频信息处理为RGB图像序列。由于这个原因,需要初步解码过程,因为视频数据通常以压缩格式存储。然而,对视频进行解码需要高计算负载和存储器使用。为了克服这个问题,我们提出了一种能够直接从压缩视频中学习的深度神经网络。我们的方法在两个公共基准测试(UCF-101和HMDB-51数据集)上进行了评估,证明了与最先进方法相当的识别性能,其优势是推理速度快2倍。

英文摘要:

Human action recognition has become one of the most active field of research in computer vision due to its wide range of applications, like surveillance, medical, industrial environments, smart homes, among others. Recently, deep learning has been successfully used to learn powerful and interpretable features for recognizing human actions in videos. Most of the existing deep learning approaches have been designed for processing video information as RGB image sequences. For this reason, a preliminary decoding process is required, since video data are often stored in a compressed format. However, a high computational load and memory usage is demanded for decoding a video. To overcome this problem, we propose a deep neural network capable of learning straight from compressed video. Our approach was evaluated on two public benchmarks, the UCF-101 and HMDB-51 datasets, demonstrating comparable recognition performance to the state-of-the-art methods, with the advantage of running up to 2 times faster in terms of inference speed.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑