search

soeaver / Parsing-R-CNN

Parsing R-CNN for Instance-Level Human Analysis
MIT License
300 stars 35 forks source link

How to get the COCO style json file for MHP or CHIP dataset? #5

Open zyf12389 opened 4 years ago

zyf12389 commented 4 years ago

annotations style in COCO must be binary. So, what is your json file generated by, is it based on the whole person or the body part? Thanks.

jiawenhao2015 commented 3 years ago

the same question~~~~thanks

jiawenhao2015 commented 3 years ago

annotations style in COCO must be binary. So, what is your json file generated by, is it based on the whole person or the body part? Thanks.

hello~do you find the way get the json anno file after 1 year?......many thanks.~~~

zyf12389 commented 3 years ago

I just added the path of the annotation to the json file in coco format.(e.g. annotation_info = { "id": annotation_id, "image_id": image_id, "bbox": bounding_box.tolist(), "parsing_file": annotation_filename, ...... })

jiawenhao2015 commented 3 years ago

I just added the path of the annotation to the json file in coco format.(e.g. annotation_info = { "id": annotation_id, "image_id": image_id, "bbox": bounding_box.tolist(), "parsing_file": annotation_filename, ...... })

many many thanks for your reply~~~~ so you mean, the json file actually contains no label info, but the really label img's path? do you change the dataset.py code? or use some other tools convert the MHP or CHIP label to coco style?

zyf12389 commented 3 years ago

Yes, the json file actually contains label annotation's path and used pycococreator.

jiawenhao2015 commented 3 years ago

Yes, the json file actually contains label annotation's path and used pycococreator.

ok ~~thanks a lot~~~i will try ~~~