当前位置:网站首页>视频异常检测数据集 (ShanghaiTech)
视频异常检测数据集 (ShanghaiTech)
2022-06-23 16:49:00 【因吉】
1 引入
源地址:https://svip-lab.github.io/dataset/campus_dataset.html
论文地址:https://openaccess.thecvf.com/content_cvpr_2018/papers/Liu_Future_Frame_Prediction_CVPR_2018_paper.pdf
数据集及代码:https://github.com/StevenLiuWen/ano_pred_cvpr2018
2 概述
异常检测模型训练的一个直接目标是可以直接应用于多视角多场景中。然而,几乎所有已有数据集都只包含使用一台固定角度相机拍摄的视频,且缺乏场景和视角的多样性。为了增加场景多样性,构建了一个新的异常检测数据集ShanghaiTech。此外,在该数据集中引入了由突然运动引起的异常,例如追逐和斗殴。这些异常是现有数据集不具备的,使得所构建的数据集更适合真实场景。
异常检测数据集总结如下:
1)CUHK Avenue:包含16个训练视频和21个测试视频,共47个异常事件,包含投掷物体、闲逛和跑步。人的大小会因为相机的位置和角度而改变;
2)Pedestrian 1 (Ped1):包含34个训练视频和36个测试视频,其中包含40个不规则事件。所有这些异常案例都与自行车、汽车等交通工具有关。
3)Pedestrian 2 (Ped2):包含16个训练视频和12个测试视频,包含12个异常事件。Ped2 的异常定义与 Ped1 相同。
4)Subway:有两类,即入口和出口。不寻常的事件包括走错方向和游荡。该数据集是在室内环境中记录的,而以上数据是在室外环境中记录。
5)ShanghaiTech:包含13个场景,具有复杂的光照条件和摄像机角度。包含130 个异常事件和超过270000个训练帧。此外,异常事件的像素级标注被给出。
3 Bib
@inproceedings{
Liu:2018:65366545,
author = {
Wen Liu and Wei Xin Luo and Dong Ze Lian and Sheng Hua Gao},
title = {
Future frame prediction for anomaly detection--a new baseline},
booktitle = {
{
IEEE} conference on Computer Vision and Pattern Recognition},
pages = {
6536--6545},
year = {
2018}
}
边栏推荐
- MySQL installation, configuration and uninstall
- How to design a seckill system - geek course notes
- QT layout manager [qvboxlayout, qhboxlayout, qgridlayout]
- 浅析3种电池容量监测方案
- Introduction to GTS Academy
- How to make validity table
- C. Set or Decrease-Educational Codeforces Round 120 (Rated for Div. 2)
- Performance test bottleneck tuning in 10 minutes! If you want to enter a large factory, you must know
- MySQL的 安裝、配置、卸載
- Reinforcement learning series (I) -- basic concepts
猜你喜欢

Ctfshow PHP features

What does the timestamp 90K mean?

hands-on-data-analysis 第二单元 第四节数据可视化

Performance test bottleneck tuning in 10 minutes! If you want to enter a large factory, you must know

Troubleshooting of datanode entering stale status

QT布局管理器【QVBoxLayout,QHBoxLayout,QGridLayout】
![[go] calling Alipay to scan code for payment in a sandbox environment](/img/d4/c6d72a697bc08f69f11121a15109b3.png)
[go] calling Alipay to scan code for payment in a sandbox environment

C # connection to database

Meituan Sanmian: how do you understand the principle of redis master-slave replication?

FPN characteristic pyramid network
随机推荐
Async/await
手机开户流程是什么?现在网上开户安全么?
FPN characteristic pyramid network
Company offensive operation guide
Query the size of each table in the database
Listen attentively and give back sincerely! Pay tribute to the best product people!
POC about secureworks' recent azure Active Directory password brute force vulnerability
MySQL事务提交流程
Analysis of three battery capacity monitoring schemes
AMQP protocol
Kerberoasting without SPN
Wechat applet: time selector for the estimated arrival date of the hotel
Meituan Sanmian: how do you understand the principle of redis master-slave replication?
浅析3种电池容量监测方案
Codeforces Round #620 (Div. 2)ABC
千呼万唤,5G双卡双通到底有多重要?
Practice sharing of chaos engineering in stability management of cloud native Middleware
Interpretation of eventbus source code
How to make validity table
C. Phoenix and Towers-Codeforces Global Round 14