search

yrcong / RelTR

RelTR: Relation Transformer for Scene Graph Generation: https://arxiv.org/abs/2201.11460v2
229 stars 45 forks source link

About OpenV6 #51

Open YaoShangJing opened 2 months ago

YaoShangJing commented 2 months ago

Why is the number of object classes in the OpenV6 dataset I processed 600, while it is mentioned as 289 object classes in the article? Were there any additional processing steps involved?

yrcong commented 2 months ago

OpenImages V6 do have more object categories. However, for the VRD tasks, many categories do not appear in the relationship annotation. For data processing, please check BGNN. I remember that I follow their setting.

YaoShangJing commented 2 months ago

非常感谢您能在百忙之中回复我的问题。我阅读了BGNN，论文中提到的目标类为301个， image 但它提供的后处理代码和处理过的标注类别好像还是600个。 image 按照您在data目录中提供的处理代码，我得出来的标注文件目标类为600个。我是不是还漏了哪步没有做呀。

yrcong commented 2 months ago

并不是所有600类都有relaitonship annotation的，举个例子，taxi在object detection task里有标注，但是并没有对应的triplet taxi-rel-obj或者sub-rel-taxi。因此就会把这种category滤掉