BrandonHanx / FAME-ViL

[CVPR 2023 (Highlight)] FAME-ViL: Multi-Tasking V+L Model for Heterogeneous Fashion Tasks
Other
51 stars 4 forks source link

What is test protocol of FashionIQ,? #8

Open BUAADreamer opened 1 year ago

BUAADreamer commented 1 year ago

❓ Questions and Help

Did you use the test protocol of original val with 3,817, 6,346 and 5,373 images or the protocol of the paper Image Search With Text Feedback by Visiolinguistic Attention Learning which is the union of the reference and target images with redundancy removed? Such two protocols may produce large difference of recall score