search

IDEA-Research / ED-Pose

[ICLR 2023] Official implementation of the paper "Explicit Box Detection Unifies End-to-End Multi-Person Pose Estimation "
Other
152 stars 10 forks source link

AR = 0.0 #16

Open cici203 opened 1 year ago

cici203 commented 1 year ago

Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets= 20 ] = 0.540 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets= 20 ] = 0.721 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets= 20 ] = 0.600 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 20 ] = 0.663 Average Recall (AR) @[ IoU=0.50 | area= all | maxDets= 20 ] = 0.829 Average Recall (AR) @[ IoU=0.75 | area= all | maxDets= 20 ] = 0.732 Average Recall (AR) @[ IoU=0.50:0.95 | type= easy | maxDets= 20 ] = 0.010 Average Recall (AR) @[ IoU=0.50:0.95 | type=medium | maxDets= 20 ] = 0.010 Average Recall (AR) @[ IoU=0.50:0.95 | type= hard | maxDets= 20 ] = 0.010

yangjie-cv commented 1 year ago

Hello, which model output is the result obtained?

youngnuaa commented 11 months ago

我也遇到相同问题，发现crowdpose数据集，AR是0，AP没有APE, APM, APH。coco数据集正常

yangjie-cv commented 11 months ago

Hi, thanks for your interest! This is a problem in CrowdPose API. You should follow https://github.com/Jeff-sjtu/CrowdPose/pull/24 to fix related python files.

cici203 commented 11 months ago

Hi, thanks for your interest! This is a problem in CrowdPose API. You should follow Jeff-sjtu/CrowdPose#24 to fix related python files.

修改后问题依然存在，只是把AR改为了AP，但是计算值为0的情况依然存在 修改前： Average Precision (AR) @[ IoU=0.50:0.95 | type= easy | maxDets= 20 ] = 0.000 Average Precision (AR) @[ IoU=0.50:0.95 | type=medium | maxDets= 20 ] = 0.000 Average Precision (AR) @[ IoU=0.50:0.95 | type= hard | maxDets= 20 ] = 0.010 修改后： Average Precision (AP) @[ IoU=0.50:0.95 | type= easy | maxDets= 20 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 | type=medium | maxDets= 20 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 | type= hard | maxDets= 20 ] = 0.010

cici203 commented 11 months ago

问题可能出在get_type_result（）