FallAction_dataset.zip

FallAction Dataset for Uncertain Action Recognition<br>Nitika Nigam, Deepali Verma, and Tanima DuttaIndian Institute of Technology (BHU), India.<br>Overview:FallAction is an uncertain action recognition dataset of realistic people falling videoscollected from YouTube. The dataset comprises 15 fall action categories, and eachcategory contains 50-100 videos. FallAction gives the diversity in terms of differentfalling actions and with the presence of noises, such as, variations in camera motions,person appearance, viewpoint, cluttered background, illumination conditions, etc. It is achallenging dataset for uncertain action recognition. Most action recognition datasetsare based on certain actions; on the contrary, FallAction aims to encourage furtherresearch into uncertain action recognition by learning and exploring new realisticuncertain fall action categories.Structure for FallAction Dataset● Data associated with each fall action category is stored in separate directories.● Each directory comprises *.mp4 files for videos.● The directory is arranged in the following structure:FallAction_datasets├── Butt_fall├── Drunk_fall├── Elderly_fall├── Escalator_fall├── Faint_fall├── False_start_fall├── Handstand_fall├── High_Heel_Break_fall├── Karate_fall├── Parkour_fall├── Pool_fall├── Side_fall├── Surfing_fall├── Trust_fall└── Yoga_fall

用于不确定动作识别（Uncertain Action Recognition）的FallAction数据集
Nitika Nigam、Deepali Verma及Tanima Dutta，印度理工学院（BHU），印度。

概述：FallAction是一款面向不确定动作识别的真实人体跌倒视频数据集，素材采集自YouTube平台。该数据集涵盖15种跌倒动作类别，每个类别包含50至100段视频。FallAction不仅具备多样化的跌倒动作类型，还融入了多种噪声干扰，例如相机运动变化、人物外观差异、拍摄视角、杂乱背景、光照条件波动等，是一个极具挑战性的不确定动作识别数据集。当前绝大多数动作识别数据集均以确定动作为研究基础，与之相反，FallAction旨在通过学习与探索新型真实不确定跌倒动作类别，推动不确定动作识别领域的进一步研究。

FallAction数据集结构
● 各跌倒动作类别的关联数据均存储于独立目录中。
● 每个目录内均包含格式为*.mp4的视频文件。
● 数据集目录结构如下：
FallAction_datasets
├── Butt_fall
├── Drunk_fall
├── Elderly_fall
├── Escalator_fall
├── Faint_fall
├── False_start_fall
├── Handstand_fall
├── High_Heel_Break_fall
├── Karate_fall
├── Parkour_fall
├── Pool_fall
├── Side_fall
├── Surfing_fall
├── Trust_fall
└── Yoga_fall