EANet
EANet copied to clipboard
EANet: Enhancing Alignment for Cross-Domain Person Re-identification
您好,我用了OPENPOSE提取了18个关键点,依次包含鼻、颈、右肩、右肘、右腕、左肩、左肘、左腕、右髋、右膝、右踝、左髋、左膝、左踝、右眼、左眼、右耳和左耳。但是您项目中应该是17个关键点存储在 im_path_to_kpt.pkl。我在自己生成pkl文件的时候,不知道您的17个关键点是哪17个,然后存放的顺序是怎么样的。感谢您的解答和帮助!

Hi kindly suggest how to generate part labels and key points for other datasets
hello sir, to visualize rank result how to provide pap mask for query and gallery image
is the python 2.7 and pytorch 1.0.0 are not supportive for tripletloss ??
"__main__", mod_spec) File "C:\Users\padmashree\anaconda3\envs\savitri\lib\runpy.py", line 85, in _run_code exec(code, run_globals) File "D:\project_dir\EANet\package\optim\eanet_trainer.py", line 135, in trainer.train_phases() File "D:\project_dir\EANet\package\optim\eanet_trainer.py", line 126, in train_phases self.train() File "D:\project_dir\EANet\package\optim\reid_trainer.py", line 338, in train self.trainer.train_one_epoch(trial_run_steps=3...
Hi, I am reproducing your paper result . when enabled ps_loss, id_loss, tri_loss as true . im getting bellow error. what could be the reason? hepl me to resolve ,...