kok电子竞技

1、The Honda/UCSD Video Database(加利福尼亚大学圣迭戈分校脸部追踪视频数据库数据摘要：The goal of the Honda/UCSD Video Database is to provide a standard video database for evaluting face tracking/recognition algorithms. Each video sequence is recorded in an indoor environment at 15 frames per second, and each lasted for at least

2、 15 seconds. The resolution of each video sequence is 640x480. Every individual is recorded in at least two video sequences. Since we believe that pose variation provides the greatest challenge to recognition, all the video sequences containsignificant 2-D (in-plane and 3-D (out-of-plane head rotati

3、ons. In each video, the person rotates and turns his/her head in his/her own preferred order and speed, and typically in about 15 seconds, the individual is able to provide a wide range of different poses. In addition, some of these sequences contain difficult events which a real-world tracker/recog

4、nizer would likely encounter, such as partial occlusion, face partly leaving the field of view, and large scale changes, etc.The Honda/UCSD Video Database contains two datasets. The first dataset is recorded by a SONY EVI-D30 camera at Honda Research Institute in 2002. It includes three different su

5、bsets, Training, Testing, and OcclusionTesting. Each subset contains 20, 42, 13 videos repectively from 20 human subjects. The second dataset is recorded by a SONY DFW-V500 camera at Computer Vision Labortory, University of California, San Diego in 2004. It includes two subsets, Training and Testing

6、 of 30 videos from another 15 different human subjects.The first dataset is used and reported by Kuang-Chih Lee, Jeffrey Ho, Ming-Hsuan Yang, David Kriegman in their paper, Visual Tracking and Recognition Using Probabilistic Appearance Manifolds, CVIU 2005 pdf (For a shorter, earlier version, see pd

7、f . The second dataset is recorded and used in our recent paper, Online Learning of Probabilistic Appearance Manifolds for Video-based Recognition and Tracking, CVPR 2005 pdf for constructing the generic face appearance manifo中文关键词：评估, 脸部追踪, 识别, 户内环境, 视频序列,英文关键词：evaluting,face tracking,recognition,i

8、ndoor environment,video sequence,数据格式：VIDEO数据用途：To provide a standard video database for evaluting facetracking/recognition algorithms.数据详细介绍：The Honda/UCSD Video DatabaseThe goal of the Honda/UCSD Video Database is to provide a standard video database for evaluting face tracking/recognition algorit

9、hms. Each video sequence is recorded in an indoor environment at 15 frames per second, and each lasted for at least 15 seconds. The resolution of each video sequence is 640x480. Every individual is recorded in at least two video sequences. Since we believe that pose variation provides the greatest c

10、hallenge to recognition, all the video sequences contain significant 2-D (in-plane and 3-D (out-of-plane head rotations. In each video, the person rotates and turns his/her head in his/her own preferred order and speed, and typically in about 15 seconds, the individual is able to provide a wide rang

11、e of different poses. In addition, some of these sequences contain difficult events which a real-world tracker/recognizer would likely encounter, such as partial occlusion, face partly leaving the field of view, and large scale changes, etc.The Honda/UCSD Video Database contains two datasets. The fi

12、rst dataset is recorded by a SONY EVI-D30 camera at Honda Research Institute in 2002. It includes three different subsets, Training, Testing, and Occlusion Testing. Each subset contains 20, 42, 13 videos repectively from 20 human subjects. The second dataset is recorded by a SONY DFW-V500 camera at

13、Computer Vision Labortory, University of California, San Diego in 2004. It includes two subsets, Training and Testing of 30 videos from another 15 different human subjects.The first dataset is used and reported by Kuang-Chih Lee, Jeffrey Ho, Ming-Hsuan Yang, David Kriegman in their paper, Visual Tra

14、cking and Recognition Using Probabilistic Appearance Manifolds, CVIU 2005 (For a shorter, earlier version, see . The second dataset is recorded and used in our recent paper, Online Learning of Probabilistic Appearance Manifolds for Video-based Recognition and Tracking, CVPR 2005 for constructingthe

15、generic face appearance manifold.The database now can be downloaded at:First DatasetSecond DatasetYou are free to use the Honda/UCSD Video Database for research purposes. All publications which uses this database to obtain results should acknowledge the use of Honda/UCSD Video Database and reference .数据预览： 点此下载完整数据集