公开数据集
数据结构 ? 1.66G
Data Structure ?
* 以上分析是由系统提取分析形成的结果,具体实际数据为准。
README.md
UCF Sports dataset consists of a set of actions collected from various sports which are typically featured on broadcast television channels such as the BBC and ESPN. The video sequences were obtained from a wide range of stock footage websites including BBC Motion gallery and GettyImages.
The dataset includes a total of 150 sequences with the resolution of 720 x 480. The collection represents a natural pool of actions featured in a wide range of scenes and viewpoints. By releasing the data set we hope to encourage further research into this class of action recognition in unconstrained environments. Since its introduction, the dataset has been used for numerous applications such as: action recognition, action localization, and saliency detection.
Dataset Actions
The dataset includes the following 10 actions. The figure above shows the a sample frame of all ten actions, along with their bounding box annotations of the humans shown in yellow.
Diving (14 videos) Golf Swing (18 videos) Kicking (20 videos) Lifting (6 videos) Riding Horse (12 videos) Running (13 videos) SkateBoarding (12 videos) Swing-Bench (20 videos) Swing-Side (13 videos) Walking (22 videos)
Dataset Summary
The following table summarizes the characteristics of the dataset.
Figure: Summary of the characteristics of UCF Sports.
Statistics
The following figure shows the distribution of the number of clips per action as the number of clips in each class is not the same.
Figure: Number of clips per action class.
The following figure illustrates the total duration of clips (blue) and the average clip length (green) for every action class. It is evident that certain actions are short in nature, such as kicking, as compared to walking or running, which are relatively longer and have more periodicity. However, it is apparent from the chart that the average duration of action clips shows great similarities across different classes. Therefore, merely considering the duration of one clip would not be enough for identifying the action.
Figure: The total time of video clips for each action class is shown in blue. Average length of clips for each action is shown in green.
Recommended Experimental Setup
Action Recognition
Leave-One-Out (LOO) cross-validation scheme: It is recommended as in [1] to test on UCF Sports by using a Leave-One-Out (LOO) cross-validation scheme. This scenario takes out one sample video for testing and trains using all of the remaining videos of an action class. This is performed for every sample video in a cyclic manner, and the overall accuracy is obtained by averaging the accuracy of all iterations.
Action Localization
Train/Test Splits: It is recommended to use train/test splits as suggested in [*]. The proposed experimental setup splits the dataset into two uneven parts: two-third of videos for training and one-third for testing. To calculate the accuracy, an intersection-over-union criterion is used to plot ROC curves with a certain overlap threshold. The intersection-over-union computes the overlap between the predicted bounding box and the ground truth, and divides it by the union of both the bounding boxes, for every frame. This value is then averaged over all frames in a video. A 20 % overlap threshold is used for this experiment. Area Under Curve (AUC) against the overlap threshold, which shows how the performance varies if the threshold is changed, is used to compute the final performance. To calculate the overlap, the ground truth bounding box per frame is provided for the dataset.
[*] Tian Lan, Yang Wang and Greg Mori, Discriminative figure-centric models for joint action localization and recognition, IEEE International Conference on Computer Vision (ICCV), 2011.
Download
The data set can be downloaded by clicking here.
Human gaze annotations can be downloaded by clicking here.
Train/Test splits for Action localization can be downloaded by clicking here.
Related Publications
If you use this data set, please cite the following papers:
- [1] Mikel D. Rodriguez, Javed Ahmed, and Mubarak Shah, Action MACH: A Spatio-temporal Maximum Average Correlation Height Filter for Action Recognition, Computer Vision and Pattern Recognition, 2008.
- [2] Khurram Soomro and Amir R. Zamir, Action Recognition in Realistic Sports Videos, Computer Vision in Sports. Springer International Publishing, 2014.
帕依提提提温馨提示
该数据集正在整理中,为您准备了其他渠道,请您使用
- 分享你的想法
全部内容
数据使用声明:
- 1、该数据来自于互联网数据采集或服务商的提供,本平台为用户提供数据集的展示与浏览。
- 2、本平台仅作为数据集的基本信息展示、包括但不限于图像、文本、视频、音频等文件类型。
- 3、数据集基本信息来自数据原地址或数据提供方提供的信息,如数据集描述中有描述差异,请以数据原地址或服务商原地址为准。
- 1、本站中的所有数据集的版权都归属于原数据发布者或数据提供方所有。
- 1、如您需要转载本站数据,请保留原数据地址及相关版权声明。
- 1、如本站中的部分数据涉及侵权展示,请及时联系本站,我们会安排进行数据下线。