In any case, I can`t detect "person" by your model

hanoonaR / object-centric-ovd

[NeurIPS 2022] Official repository of paper titled "Bridging the Gap between Object and Image-level Representations for Open-Vocabulary Detection".

Apache License 2.0

284 stars 19 forks source link

Closed sixone-Jiang closed 1 year ago

sixone-Jiang commented 1 year ago

Thanks for your excellent work, but I meet some trouble. In any case, I can`t detect "person" by your model. (use coco or lvis, even self define)