InfAR
收藏资源简介:
InfAR数据集是由西安电子科技大学电信工程学院的研究团队创建的,包含600个由红外热成像摄像机捕捉的视频序列。该数据集主要用于红外人行动作识别研究,涵盖了12种不同的动作类别,每类包含50个视频片段,每个片段平均持续4秒,帧率为25fps,分辨率为293x256。数据集中的视频包含单人或多人的动作,部分涉及多人互动。InfAR数据集的创建旨在解决红外人行动作数据有限的问题,通过提供丰富的红外动作数据,推动红外人行动作识别技术的发展。
The InfAR Dataset was developed by a research team from the School of Telecommunications Engineering, Xidian University. It contains 600 video sequences captured using infrared thermal imaging cameras. This dataset is primarily intended for research on infrared human action recognition, encompassing 12 distinct action categories. Each category includes 50 video clips, with each clip having an average duration of 4 seconds, a frame rate of 25 fps, and a resolution of 293×256. The videos in the dataset feature actions performed by single or multiple individuals, with some segments involving multi-person interactions. The creation of the InfAR Dataset aims to address the scarcity of existing infrared human action data, and to advance the development of infrared human action recognition technologies by providing a rich collection of infrared action data.

- 1Transferable Feature Representation for Visible-to-Infrared Cross-Dataset Human Action Recognition西安电子科技大学电信工程学院 · 2019年



