ray-tju/mfirrn

This repository contains the source codes for the paper "MULTI-GRANULARITY FEATURE INTERACTION AND RELATION REASONING FOR 3D DENSE ALIGNMENT AND FACE RECONSTRUCTION (ICASSP 2021)"

mfirrn's Issues

About weights file

Hi! Thanks for your great job!

I want to get 3D vertices results of AFLW2000-3D with your project, and I find there is no test file and your trained weights file. Maybe I got something wrong? If not, would you mind release your model weights for the test? Hope for your reply! Thx!

One more issue about the model

The file https://github.com/leilimaster/MFIRRN/blob/main/model/Mfirrn.py in ./model has one module named 'attention'. I tried to add another module like:

import torch.nn as nn
class SELayer(nn.Module):
def init(self, channel, reduction=16):
super(SELayer, self).init()
self.avg_pool = nn.AdaptiveAvgPool2d(1)
self.fc = nn.Sequential(
nn.Linear(channel, channel // reduction, bias=False),
nn.ReLU(inplace=True),
nn.Linear(channel // reduction, channel, bias=False),
nn.Sigmoid()
)

def forward(self, x):
    b, c, _, _ = x.size()
    y = self.avg_pool(x).view(b, c)
    y = self.fc(y).view(b, c, 1, 1)
    return x * y.expand_as(x)

as a .py file and then 'import attention' in Mfirrn.py.

But, when I run the benchmark.py, it can't work.

error:
RuntimeError: Error(s) in loading state_dict for LLNet:
Unexpected key(s) in state_dict: "attention.fc.0.weight", "attention.fc.2.weight".

Could you give some guidance?

Cannot find the paper link

Could you please update the paper link?

Could you release the pretrained model for evaluation?

Please...