search

Wanggcong / Spatial-Temporal-Re-identification

[AAAI 2019] Spatial Temporal Re-identification
MIT License
384 stars 78 forks source link

Why two features are extarcted for one image? [feature extraction in test.py] #33

Open KrystalCWT opened 4 years ago

KrystalCWT commented 4 years ago

In your code: for i in range(2): if(i==1): img = fliplr(img) input_img = Variable(img.cuda()) outputs = model(input_img) f = outputs.data.cpu() ff = ff+f

Why two features are extarcted for one image?

Wanggcong commented 4 years ago

This is a basic setting in the current state-of-the-art methods. An image and its flipped image are used for feature extraction.