Open pikerbright opened 5 years ago
I found that w-net network is actually the same as SFANet in paper "Dual Path Multi-Scale Fusion Networks with Attention for Crowd Counting". Some hyper parameters and training details are also the same.
Can someone help to review these papers?
Creepy, based on the arXiv-ed time stamps, SFANet seems to have appeared earlier.
I found that w-net network is actually the same as SFANet in paper "Dual Path Multi-Scale Fusion Networks with Attention for Crowd Counting". Some hyper parameters and training details are also the same.
Can someone help to review these papers?