Dataset problem: Why Duke dataset is so big? Can you provide how to create Occluded dataset??

nickhuang1996 / HJL-re-id

A awesome project for person re-id. There are re-implements of PCB, MGN, MDRS(ours), PGFA, Pyramidal and HOReID.

MIT License

53 stars 17 forks source link

Dataset problem: Why Duke dataset is so big? Can you provide how to create Occluded dataset?? #5

Open zx081325 opened 4 years ago

zx081325 commented 4 years ago

Hi, nickhuang1996, I have a problem that when I download the Duke dataset, i find that it took a long time. I want to separate the occluded dataset from the duke dataset. But I consider that if there is any problem for moving location of it. Please suggest me how to do with it.

nickhuang1996 commented 4 years ago

@zx081325 The location is flexable. If you want to separate the occluded dataset, you can study the structure of the duke dataset and write code for this.