Face-in-Action

Face-in-Action数据集是一个具有成像变化的面部视频数据集，它涵盖了来自180个参与者的20秒面部视频数据，主要用于模仿护照控制场景，并且相关视频由6个同时摄像机从3个不同角度捕获 (每个角度的8毫米和4毫米的焦距)，在密闭的室内环境和开放的室外环境中收集图像。
此数据集由卡内基梅隆大学2010年发布。

The Face-in-Action Dataset is a facial video dataset with imaging variations. It contains 20-second facial video data from 180 participants, which is primarily designed to simulate passport control scenarios. The related videos were captured by 6 simultaneous cameras across 3 distinct angles, with focal lengths of 8 mm and 4 mm for each angle, and the images were collected in both enclosed indoor environments and open outdoor environments. This dataset was released by Carnegie Mellon University in 2010.

搜集汇总

背景与挑战

背景概述

Face-in-Action是一个面部视频数据集，包含180名参与者的20秒视频，用于模拟护照控制场景。视频由6个摄像机从3个不同角度拍摄，涵盖室内和室外环境，由卡内基梅隆大学于2010年发布。