FWNet architecture using pytorch

model.py

# -*- coding: utf-8 -*-
"""
Created on Tue Dec 17 08:04:29 2019

@author: shino
"""

from __future__ import print_function
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.utils.data
from torch.autograd import Variable
import numpy as np
import torch.nn.functional as F


class STN3d(nn.Module):
    def __init__(self):
        super(STN3d, self).__init__()
        self.conv1 = torch.nn.Conv1d(163, 128, 1)
        self.conv2 = torch.nn.Conv1d(128, 256, 1)
        self.conv3 = torch.nn.Conv1d(256, 1024, 1)
        self.fc1 = nn.Linear(1024, 512)
        self.fc2 = nn.Linear(512, 256)
        self.fc3 = nn.Linear(256, 9)
        self.relu = nn.ReLU()

        self.bn1 = nn.BatchNorm1d(128)
        self.bn2 = nn.BatchNorm1d(256)
        self.bn3 = nn.BatchNorm1d(1024)
        self.bn4 = nn.BatchNorm1d(512)
        self.bn5 = nn.BatchNorm1d(256)


    def forward(self, x):
        batchsize = x.size()[0]
        x = F.relu(self.bn1(self.conv1(x)))
        x = F.relu(self.bn2(self.conv2(x)))
        x = F.relu(self.bn3(self.conv3(x)))
        x = torch.max(x, 2, keepdim=True)[0]
        x = x.view(-1, 1024)

        x = F.relu(self.bn4(self.fc1(x)))
        x = F.relu(self.bn5(self.fc2(x)))
        x = self.fc3(x)

        iden = Variable(torch.from_numpy(np.array([1,0,0,0,1,0,0,0,1]).astype(np.float32))).view(1,9).repeat(batchsize,1)
        if x.is_cuda:
            iden = iden.cuda()
        x = x + iden
        x = x.view(-1, 3, 3)
        return x


class STNkd(nn.Module):
    def __init__(self, k=163):
        super(STNkd, self).__init__()
        self.conv1 = torch.nn.Conv1d(k, 128, 1)
        self.conv2 = torch.nn.Conv1d(128, 256, 1)
        self.conv3 = torch.nn.Conv1d(256, 1024, 1)
        self.fc1 = nn.Linear(1024, 512)
        self.fc2 = nn.Linear(512, 256)
        self.fc3 = nn.Linear(256, k*k)
        self.relu = nn.ReLU()

        self.bn1 = nn.BatchNorm1d(128)
        self.bn2 = nn.BatchNorm1d(256)
        self.bn3 = nn.BatchNorm1d(1024)
        self.bn4 = nn.BatchNorm1d(512)
        self.bn5 = nn.BatchNorm1d(256)

        self.k = k

    def forward(self, x):
        batchsize = x.size()[0]
        #print('x = '+str(x.size()))
        #print('batchsize = '+str(batchsize))
        x = F.relu(self.bn1(self.conv1(x)))
        x = F.relu(self.bn2(self.conv2(x)))
        x = F.relu(self.bn3(self.conv3(x)))
        x = torch.max(x, 2, keepdim=True)[0]
        x = x.view(-1, 1024)

        x = F.relu(self.bn4(self.fc1(x)))
        x = F.relu(self.bn5(self.fc2(x)))
        x = self.fc3(x)

        iden = Variable(torch.from_numpy(np.eye(self.k).flatten().astype(np.float32))).view(1,self.k*self.k).repeat(batchsize,1)
        if x.is_cuda:
            iden = iden.cuda()
        x = x + iden
        x = x.view(-1, self.k, self.k)
        return x

class PointNetfeat(nn.Module):
    def __init__(self, global_feat = True, feature_transform = False, isVAE=False, istrain = True):
        super(PointNetfeat, self).__init__()
        self.stn = STNkd()
        self.conv1 = torch.nn.Conv1d(163, 128, 1)
        self.conv2 = torch.nn.Conv1d(128, 256, 1)
        self.conv3 = torch.nn.Conv1d(256, 1024, 1)
        self.bn1 = nn.BatchNorm1d(128)
        self.bn2 = nn.BatchNorm1d(256)
        self.bn3 = nn.BatchNorm1d(1024)
        self.global_feat = global_feat
        self.feature_transform = feature_transform
        if self.feature_transform:
            self.fstn = STNkd(k=64)
        self.isVAE = isVAE
        self.istrain = istrain

    def forward(self, x):
        n_pts = x.size()[2]
        #print('n_pts = '+str(n_pts))
        trans = self.stn(x)
        x = x.transpose(2, 1)
        x = torch.bmm(x, trans)
        x = x.transpose(2, 1)
        x = F.relu(self.bn1(self.conv1(x)))

        if self.feature_transform:
            trans_feat = self.fstn(x)
            x = x.transpose(2,1)
            x = torch.bmm(x, trans_feat)
            x = x.transpose(2,1)
        else:
            trans_feat = None

        pointfeat = x
        x = F.relu(self.bn2(self.conv2(x)))
        x = self.bn3(self.conv3(x))
        if(self.isVAE == True):
             if(self.istrain == True):
                 rand = torch.rand(x.size()).cuda()
                 #print(rand.size())
                 x = x + rand 
                 #print(x.size())
             elif(self.istrain == False):
                 x = x
              
        x = torch.max(x, 2, keepdim=True)[0]
        x = x.view(-1, 1024)


        if self.global_feat:
            return x, trans, trans_feat
        else:
            x = x.view(-1, 1024, 1).repeat(1, 1, n_pts)
            return torch.cat([x, pointfeat], 1), trans, trans_feat

class PointNetCls(nn.Module):
    def __init__(self, k=6, feature_transform=False):
        super(PointNetCls, self).__init__()
        self.feature_transform = feature_transform
        self.feat = PointNetfeat(global_feat=True, feature_transform=feature_transform)
        self.fc1 = nn.Linear(1024, 512)
        self.fc2 = nn.Linear(512, 256)
        self.fc3 = nn.Linear(256, k)
        self.dropout = nn.Dropout(p=0.3)
        self.bn1 = nn.BatchNorm1d(512)
        self.bn2 = nn.BatchNorm1d(256)
        self.relu = nn.ReLU()

    def forward(self, x):
        x, trans, trans_feat = self.feat(x)
        x = F.relu(self.bn1(self.fc1(x)))
        x = F.relu(self.bn2(self.dropout(self.fc2(x))))
        x = self.fc3(x)
        return F.log_softmax(x, dim=1), trans, trans_feat


class PointNetDenseCls(nn.Module):
    def __init__(self, k = 163, feature_transform=False, isVAE = False, istrain = True):
        super(PointNetDenseCls, self).__init__()
        self.k = k
        self.feature_transform=feature_transform
        self.feat = PointNetfeat(global_feat=False, feature_transform=feature_transform,isVAE=isVAE, istrain = istrain)
        self.conv1 = torch.nn.Conv1d(1152, 512, 1)
        self.conv2 = torch.nn.Conv1d(512, 256, 1)
        self.conv3 = torch.nn.Conv1d(256, 128, 1)
        self.conv4 = torch.nn.Conv1d(128, k, 1)
        self.conv4_1 = torch.nn.Conv1d(128, 3, 1)
        self.conv4_2 = torch.nn.Conv1d(128, 163, 1)
        self.bn1 = nn.BatchNorm1d(512)
        self.bn2 = nn.BatchNorm1d(256)
        self.bn3 = nn.BatchNorm1d(128)

    def forward(self, x):
        batchsize = x.size()[0]
        #print('batch size = '+str(batchsize))
        n_pts = x.size()[2]
        #print('n_pts = '+str(n_pts))
        bottleneck, trans, trans_feat = self.feat(x)
        x = F.relu(self.bn1(self.conv1(bottleneck)))
        x = F.relu(self.bn2(self.conv2(x)))
        x = F.relu(self.bn3(self.conv3(x)))
        x = self.conv4(x)
        #x_1 = self.conv4_1(x)
        #x_2 = self.conv4_2(x)
        #x = x.transpose(2,1).contiguous()
        #x = F.linear(x.view(-1,self.k), dim=-1)
        #x = x.view(batchsize, n_pts, self.k)
        #return x_1, x_2, trans, trans_feat, bottleneck
        return x, trans, trans_feat, bottleneck

def feature_transform_regularizer(trans):
    d = trans.size()[1]
    batchsize = trans.size()[0]
    I = torch.eye(d)[None, :, :]
    if trans.is_cuda:
        I = I.cuda()
    loss = torch.mean(torch.norm(torch.bmm(trans, trans.transpose(2,1)) - I, dim=(1,2)))
    return loss

def mse(pred,gt):
    loss = torch.mean(torch.norm(torch.bmm(trans, trans.transpose(2,1)) - I, dim=(1,2)))
    return loss

if __name__ == '__main__':
    sim_data = Variable(torch.rand(32,62,2048))
    print(sim_data.size())
    pointfeat = PointNetfeat(global_feat=True)
    out, _, _ = pointfeat(sim_data)
    print('global feat', out.size())

    pointfeat = PointNetfeat(global_feat=False)
    out, _, _ = pointfeat(sim_data)
    print('point feat', out.size())



    seg = PointNetDenseCls(k = 163)
    out, _, _ = seg(sim_data)
    print('seg', out.size())