about predict.py and evl_yab.py

[zhuyazhi]

1. 您好，请问predict.py和evl_yab.py这两个.py都需要运行才是测试吗？ 如果要进行测试，需要运行两个.py吗?

[wenbowen123]

Hi, predict.py is used to predict pose results. eval_ycb.py will then evaluate its saved poses and get the scores on the benchmark.

[zhuyazhi]

thank you`

1. 我运行predict.py的结果是adi_auc =97.87，但是运行eval_ycb.py的结果是 add: 97.22709481181084 adi: 97.83757093086494 这两个adi有什么关系呢？
2. 我发现在eval_ycb.py中，以下代码是用于读取pred_pose的，但是默认的输出是00000.txt和00000gt.txt在一个路径下，所以len(pose_files) = 4016 。而我将pred_txt与gt_txt分别存储在out_dir/pred/和out_dir/gt/下，所以pose_files应该只读取pred_pose（00000.txt ... 02007.txt），您认为这样是对的吗？

def eval_one_class(args): pose_files = sorted(glob.glob(args.res_dir+'pred/'+'*.txt',recursive=True)) print(len(pose_files)) #2008

3. 我注意到并不是所有的pred_pose和对应的gt_pose加入add计算，而是在关键帧里面的才计算，请问我的理解对吗？为什么要这样呢？

   seq_frame_str = '%04d/%06d'%(seq_id,frame_id)

   print(seq_frame_str)

   if seq_frame_str not in keyframes: continue pred = np.loadtxt(pose_file) gt_file = '{}/data_organized/%04d/pose_gt/{}/%06d.txt'.format(args.ycb_dir,args.class_id)%(seq_id,frame_id)

   print(gt_file)

我注释了这两句判断，最终的结果如下：

if seq_frame_str not in keyframes:

  #continue

args.class_id: 12 021_bleach_cleanser add: 97.36765553284819 adi: 97.88892223750305 希望得到您回复

[johnbhlm]

I think that different CAD models lead to different results. In predict.py file, you may be using YCB_traindata->data_info.yml->textured.ply as CAD model，but in eval_ycb.py，you may be using *points.xyz as CAD model. So when calculating ADD and ADD-S will be different.

[zhuyazhi]

thank you for your suggestion~

[wenbowen123]

@johnbhlm says is correct.

@zhuyazhi , evaluating only on keyframes is the rule of YCB-Video benchmark. You can check PoseCNN for more details on the evaluation protocol.

[zhuyazhi]

ok, thanks for your reply。

[jinzhiyang1]

ok, thanks for your reply。 大佬你好，我最近也在看这个项目，但是在eval的时候出现了一些错误，你能分享一下eval_ycbineoat.py的配置路径么，万分感谢