About ShanghaiTech Campus dataset (Anomaly Detection)

[Markovcom]

Hello, there are 130 anomaly events in this dataset introduced by your paper. What are these anomaly types specifically?

[zachluo]

semantic_annotation.zip
Hi @Markovcom ,
I have uploaded two semantic annotations of ShanghaiTech and Avenue.

[Markovcom]

@zachluo Thank you very much，do you have similar annotation files for UCSD Ped1 or Ped2?

[zachluo]

@Markovcom I don't have them for Ped1 and Ped2 since we only evaluate our method in IJCAI on ShanghaiTech and Avenue.

[Markovcom]

@zachluo Thank you all the same.

[Markovcom]

@zachluo Can you explain the format of the annotation file？

[zachluo]

@Markovcom '0001' is the video id and in this video there are two anomalies. One is chasing at Frame #[1, 169] U [172, 177] and another one is jumping at Frame #[385, 385] U [387, 500].

[Markovcom]

@zachluo Thank you very much.

[youngupup]

@zachluo 你好请问shanghaitech数据集测试集的时长是多长？