This website requires JavaScript.

Open-Source Tools for Behavioral Video Analysis: Setup, Methods, and Development

Kevin LuxemJennifer J. SunSean P. Bradley ...+4 Mark Laubach
Apr 2022
摘要
Recently developed methods for video analysis, especially models for poseestimation and behavior classification, are transforming behavioralquantification to be more precise, scalable, and reproducible in fields such asneuroscience and ethology. These tools overcome long-standing limitations ofmanual scoring of video frames and traditional "center of mass" trackingalgorithms to enable video analysis at scale. The expansion of open-sourcetools for video acquisition and analysis has led to new experimental approachesto understand behavior. Here, we review currently available open source toolsfor video analysis, how to set them up in a lab that is new to video recordingmethods, and some issues that should be addressed by developers and advancedusers, including the need to openly share datasets and code, how to comparealgorithms and their parameters, and the need for documentation andcommunity-wide standards. We hope to encourage more widespread use andcontinued development of the tools. They have tremendous potential foraccelerating scientific progress for understanding the brain and behavior.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答