xushilin1 / FashionFormer

Code for our ECCV-2022 work: Fashionformer A simple, effective and unified baseline for human fashion segmentation and recognition
Apache License 2.0
91 stars 17 forks source link

Interpretation of the result in image_demo.py #22

Open Liumucan opened 2 months ago

Liumucan commented 2 months ago

Thank you for your work and for sharing the code. I'm confused about the result variable in the image_demo.py script. I understand that the annotations contain 46 categories and 294 attributes. However, I'm wondering why the third entry in the result presents an array structure like 46x1x294. Shouldn't there be a single array of size 1x294 to represent the probabilities of the different attributes? How should I interpret the interactions between the categories and the attributes? Thank you.