diff --git a/.gitignore b/.gitignore index 81e909a..b3b71af 100644 --- a/.gitignore +++ b/.gitignore @@ -5,3 +5,4 @@ *.pyc gimpenv/ weights/ +output/ \ No newline at end of file diff --git a/gimp-plugins/RIFE/LICENSE b/gimp-plugins/RIFE/LICENSE new file mode 100755 index 0000000..4cb767a --- /dev/null +++ b/gimp-plugins/RIFE/LICENSE @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2020 hzwer + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/gimp-plugins/RIFE/model/IFNet.py b/gimp-plugins/RIFE/model/IFNet.py new file mode 100755 index 0000000..b6c93f0 --- /dev/null +++ b/gimp-plugins/RIFE/model/IFNet.py @@ -0,0 +1,117 @@ +import torch +import numpy as np +import torch.nn as nn +import torch.nn.functional as F +from model.warplayer import warp + + +# device = torch.device("cuda" if torch.cuda.is_available() else "cpu") + +def conv_wo_act(in_planes, out_planes, kernel_size=3, stride=1, padding=1, dilation=1): + return nn.Sequential( + nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, + padding=padding, dilation=dilation, bias=False), + nn.BatchNorm2d(out_planes), + ) + + +def conv(in_planes, out_planes, kernel_size=3, stride=1, padding=1, dilation=1): + return nn.Sequential( + nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, + padding=padding, dilation=dilation, bias=False), + nn.BatchNorm2d(out_planes), + nn.PReLU(out_planes) + ) + + +class ResBlock(nn.Module): + def __init__(self, in_planes, out_planes, stride=1): + super(ResBlock, self).__init__() + if in_planes == out_planes and stride == 1: + self.conv0 = nn.Identity() + else: + self.conv0 = nn.Conv2d(in_planes, out_planes, + 3, stride, 1, bias=False) + self.conv1 = conv(in_planes, out_planes, 3, stride, 1) + self.conv2 = conv_wo_act(out_planes, out_planes, 3, 1, 1) + self.relu1 = nn.PReLU(1) + self.relu2 = nn.PReLU(out_planes) + self.fc1 = nn.Conv2d(out_planes, 16, kernel_size=1, bias=False) + self.fc2 = nn.Conv2d(16, out_planes, kernel_size=1, bias=False) + + def forward(self, x): + y = self.conv0(x) + x = self.conv1(x) + x = self.conv2(x) + w = x.mean(3, True).mean(2, True) + w = self.relu1(self.fc1(w)) + w = torch.sigmoid(self.fc2(w)) + x = self.relu2(x * w + y) + return x + + +class IFBlock(nn.Module): + def __init__(self, in_planes, scale=1, c=64): + super(IFBlock, self).__init__() + self.scale = scale + self.conv0 = conv(in_planes, c, 3, 2, 1) + self.res0 = ResBlock(c, c) + self.res1 = ResBlock(c, c) + self.res2 = ResBlock(c, c) + self.res3 = ResBlock(c, c) + self.res4 = ResBlock(c, c) + self.res5 = ResBlock(c, c) + self.conv1 = nn.Conv2d(c, 8, 3, 1, 1) + self.up = nn.PixelShuffle(2) + + def forward(self, x): + if self.scale != 1: + x = F.interpolate(x, scale_factor=1. / self.scale, mode="bilinear", + align_corners=False) + x = self.conv0(x) + x = self.res0(x) + x = self.res1(x) + x = self.res2(x) + x = self.res3(x) + x = self.res4(x) + x = self.res5(x) + x = self.conv1(x) + flow = self.up(x) + if self.scale != 1: + flow = F.interpolate(flow, scale_factor=self.scale, mode="bilinear", + align_corners=False) + return flow + + +class IFNet(nn.Module): + def __init__(self, cFlag): + super(IFNet, self).__init__() + self.block0 = IFBlock(6, scale=4, c=192) + self.block1 = IFBlock(8, scale=2, c=128) + self.block2 = IFBlock(8, scale=1, c=64) + self.cFlag = cFlag + + def forward(self, x): + x = F.interpolate(x, scale_factor=0.5, mode="bilinear", + align_corners=False) + flow0 = self.block0(x) + F1 = flow0 + warped_img0 = warp(x[:, :3], F1, self.cFlag) + warped_img1 = warp(x[:, 3:], -F1, self.cFlag) + flow1 = self.block1(torch.cat((warped_img0, warped_img1, F1), 1)) + F2 = (flow0 + flow1) + warped_img0 = warp(x[:, :3], F2, self.cFlag) + warped_img1 = warp(x[:, 3:], -F2, self.cFlag) + flow2 = self.block2(torch.cat((warped_img0, warped_img1, F2), 1)) + F3 = (flow0 + flow1 + flow2) + return F3, [F1, F2, F3] + + +if __name__ == '__main__': + img0 = torch.zeros(3, 3, 256, 256).float().to(device) + img1 = torch.tensor(np.random.normal( + 0, 1, (3, 3, 256, 256))).float().to(device) + imgs = torch.cat((img0, img1), 1) + flownet = IFNet() + flow, _ = flownet(imgs) + print(flow.shape) diff --git a/gimp-plugins/RIFE/model/IFNet2F.py b/gimp-plugins/RIFE/model/IFNet2F.py new file mode 100755 index 0000000..b31a1b8 --- /dev/null +++ b/gimp-plugins/RIFE/model/IFNet2F.py @@ -0,0 +1,115 @@ +import torch +import numpy as np +import torch.nn as nn +import torch.nn.functional as F +from model.warplayer import warp + + +device = torch.device("cuda" if torch.cuda.is_available() else "cpu") + +def conv_wo_act(in_planes, out_planes, kernel_size=3, stride=1, padding=1, dilation=1): + return nn.Sequential( + nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, + padding=padding, dilation=dilation, bias=False), + nn.BatchNorm2d(out_planes), + ) + + +def conv(in_planes, out_planes, kernel_size=3, stride=1, padding=1, dilation=1): + return nn.Sequential( + nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, + padding=padding, dilation=dilation, bias=False), + nn.BatchNorm2d(out_planes), + nn.PReLU(out_planes) + ) + + +class ResBlock(nn.Module): + def __init__(self, in_planes, out_planes, stride=1): + super(ResBlock, self).__init__() + if in_planes == out_planes and stride == 1: + self.conv0 = nn.Identity() + else: + self.conv0 = nn.Conv2d(in_planes, out_planes, + 3, stride, 1, bias=False) + self.conv1 = conv(in_planes, out_planes, 3, stride, 1) + self.conv2 = conv_wo_act(out_planes, out_planes, 3, 1, 1) + self.relu1 = nn.PReLU(1) + self.relu2 = nn.PReLU(out_planes) + self.fc1 = nn.Conv2d(out_planes, 16, kernel_size=1, bias=False) + self.fc2 = nn.Conv2d(16, out_planes, kernel_size=1, bias=False) + + def forward(self, x): + y = self.conv0(x) + x = self.conv1(x) + x = self.conv2(x) + w = x.mean(3, True).mean(2, True) + w = self.relu1(self.fc1(w)) + w = torch.sigmoid(self.fc2(w)) + x = self.relu2(x * w + y) + return x + + +class IFBlock(nn.Module): + def __init__(self, in_planes, scale=1, c=64): + super(IFBlock, self).__init__() + self.scale = scale + self.conv0 = conv(in_planes, c, 3, 1, 1) + self.res0 = ResBlock(c, c) + self.res1 = ResBlock(c, c) + self.res2 = ResBlock(c, c) + self.res3 = ResBlock(c, c) + self.res4 = ResBlock(c, c) + self.res5 = ResBlock(c, c) + self.conv1 = nn.Conv2d(c, 2, 3, 1, 1) + self.up = nn.PixelShuffle(2) + + def forward(self, x): + if self.scale != 1: + x = F.interpolate(x, scale_factor=1. / self.scale, mode="bilinear", + align_corners=False) + x = self.conv0(x) + x = self.res0(x) + x = self.res1(x) + x = self.res2(x) + x = self.res3(x) + x = self.res4(x) + x = self.res5(x) + x = self.conv1(x) + flow = x # self.up(x) + if self.scale != 1: + flow = F.interpolate(flow, scale_factor=self.scale, mode="bilinear", + align_corners=False) + return flow + + +class IFNet(nn.Module): + def __init__(self): + super(IFNet, self).__init__() + self.block0 = IFBlock(6, scale=4, c=192) + self.block1 = IFBlock(8, scale=2, c=128) + self.block2 = IFBlock(8, scale=1, c=64) + + def forward(self, x): + x = F.interpolate(x, scale_factor=0.5, mode="bilinear", + align_corners=False) + flow0 = self.block0(x) + F1 = flow0 + warped_img0 = warp(x[:, :3], F1) + warped_img1 = warp(x[:, 3:], -F1) + flow1 = self.block1(torch.cat((warped_img0, warped_img1, F1), 1)) + F2 = (flow0 + flow1) + warped_img0 = warp(x[:, :3], F2) + warped_img1 = warp(x[:, 3:], -F2) + flow2 = self.block2(torch.cat((warped_img0, warped_img1, F2), 1)) + F3 = (flow0 + flow1 + flow2) + return F3, [F1, F2, F3] + +if __name__ == '__main__': + img0 = torch.zeros(3, 3, 256, 256).float().to(device) + img1 = torch.tensor(np.random.normal( + 0, 1, (3, 3, 256, 256))).float().to(device) + imgs = torch.cat((img0, img1), 1) + flownet = IFNet() + flow, _ = flownet(imgs) + print(flow.shape) diff --git a/gimp-plugins/RIFE/model/RIFE.py b/gimp-plugins/RIFE/model/RIFE.py new file mode 100755 index 0000000..99fa071 --- /dev/null +++ b/gimp-plugins/RIFE/model/RIFE.py @@ -0,0 +1,262 @@ +import torch +import torch.nn as nn +import numpy as np +from torch.optim import AdamW +import torch.optim as optim +import itertools +from model.warplayer import warp +from torch.nn.parallel import DistributedDataParallel as DDP +from model.IFNet import * +import torch.nn.functional as F +from model.loss import * + + +def conv(in_planes, out_planes, kernel_size=3, stride=1, padding=1, dilation=1): + return nn.Sequential( + nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, + padding=padding, dilation=dilation, bias=True), + nn.PReLU(out_planes) + ) + + +def deconv(in_planes, out_planes, kernel_size=4, stride=2, padding=1): + return nn.Sequential( + torch.nn.ConvTranspose2d(in_channels=in_planes, out_channels=out_planes, + kernel_size=4, stride=2, padding=1, bias=True), + nn.PReLU(out_planes) + ) + + +def conv_woact(in_planes, out_planes, kernel_size=3, stride=1, padding=1, dilation=1): + return nn.Sequential( + nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, + padding=padding, dilation=dilation, bias=True), + ) + + +class ResBlock(nn.Module): + def __init__(self, in_planes, out_planes, stride=2): + super(ResBlock, self).__init__() + if in_planes == out_planes and stride == 1: + self.conv0 = nn.Identity() + else: + self.conv0 = nn.Conv2d(in_planes, out_planes, + 3, stride, 1, bias=False) + self.conv1 = conv(in_planes, out_planes, 3, stride, 1) + self.conv2 = conv_woact(out_planes, out_planes, 3, 1, 1) + self.relu1 = nn.PReLU(1) + self.relu2 = nn.PReLU(out_planes) + self.fc1 = nn.Conv2d(out_planes, 16, kernel_size=1, bias=False) + self.fc2 = nn.Conv2d(16, out_planes, kernel_size=1, bias=False) + + def forward(self, x): + y = self.conv0(x) + x = self.conv1(x) + x = self.conv2(x) + w = x.mean(3, True).mean(2, True) + w = self.relu1(self.fc1(w)) + w = torch.sigmoid(self.fc2(w)) + x = self.relu2(x * w + y) + return x + + +c = 16 + + +class ContextNet(nn.Module): + def __init__(self, cFlag): + super(ContextNet, self).__init__() + self.conv1 = ResBlock(3, c) + self.conv2 = ResBlock(c, 2 * c) + self.conv3 = ResBlock(2 * c, 4 * c) + self.conv4 = ResBlock(4 * c, 8 * c) + self.cFlag = cFlag + + def forward(self, x, flow): + x = self.conv1(x) + f1 = warp(x, flow, self.cFlag) + x = self.conv2(x) + flow = F.interpolate(flow, scale_factor=0.5, mode="bilinear", + align_corners=False) * 0.5 + f2 = warp(x, flow, self.cFlag) + x = self.conv3(x) + flow = F.interpolate(flow, scale_factor=0.5, mode="bilinear", + align_corners=False) * 0.5 + f3 = warp(x, flow, self.cFlag) + x = self.conv4(x) + flow = F.interpolate(flow, scale_factor=0.5, mode="bilinear", + align_corners=False) * 0.5 + f4 = warp(x, flow, self.cFlag) + return [f1, f2, f3, f4] + + +class FusionNet(nn.Module): + def __init__(self, cFlag): + super(FusionNet, self).__init__() + self.down0 = ResBlock(8, 2 * c) + self.down1 = ResBlock(4 * c, 4 * c) + self.down2 = ResBlock(8 * c, 8 * c) + self.down3 = ResBlock(16 * c, 16 * c) + self.up0 = deconv(32 * c, 8 * c) + self.up1 = deconv(16 * c, 4 * c) + self.up2 = deconv(8 * c, 2 * c) + self.up3 = deconv(4 * c, c) + self.conv = nn.Conv2d(c, 4, 3, 1, 1) + self.cFlag = cFlag + + def forward(self, img0, img1, flow, c0, c1, flow_gt): + warped_img0 = warp(img0, flow, self.cFlag) + warped_img1 = warp(img1, -flow, self.cFlag) + if flow_gt == None: + warped_img0_gt, warped_img1_gt = None, None + else: + warped_img0_gt = warp(img0, flow_gt[:, :2]) + warped_img1_gt = warp(img1, flow_gt[:, 2:4]) + s0 = self.down0(torch.cat((warped_img0, warped_img1, flow), 1)) + s1 = self.down1(torch.cat((s0, c0[0], c1[0]), 1)) + s2 = self.down2(torch.cat((s1, c0[1], c1[1]), 1)) + s3 = self.down3(torch.cat((s2, c0[2], c1[2]), 1)) + x = self.up0(torch.cat((s3, c0[3], c1[3]), 1)) + x = self.up1(torch.cat((x, s2), 1)) + x = self.up2(torch.cat((x, s1), 1)) + x = self.up3(torch.cat((x, s0), 1)) + x = self.conv(x) + return x, warped_img0, warped_img1, warped_img0_gt, warped_img1_gt + + +class Model: + def __init__(self, c_flag, local_rank=-1): + self.flownet = IFNet(c_flag) + self.contextnet = ContextNet(c_flag) + self.fusionnet = FusionNet(c_flag) + self.device(c_flag) + self.optimG = AdamW(itertools.chain( + self.flownet.parameters(), + self.contextnet.parameters(), + self.fusionnet.parameters()), lr=1e-6, weight_decay=1e-5) + self.schedulerG = optim.lr_scheduler.CyclicLR( + self.optimG, base_lr=1e-6, max_lr=1e-3, step_size_up=8000, cycle_momentum=False) + self.epe = EPE() + self.ter = Ternary(c_flag) + self.sobel = SOBEL(c_flag) + if local_rank != -1: + self.flownet = DDP(self.flownet, device_ids=[ + local_rank], output_device=local_rank) + self.contextnet = DDP(self.contextnet, device_ids=[ + local_rank], output_device=local_rank) + self.fusionnet = DDP(self.fusionnet, device_ids=[ + local_rank], output_device=local_rank) + + def train(self): + self.flownet.train() + self.contextnet.train() + self.fusionnet.train() + + def eval(self): + self.flownet.eval() + self.contextnet.eval() + self.fusionnet.eval() + + def device(self, c_flag): + if torch.cuda.is_available() and not c_flag: + device = torch.device("cuda") + else: + device = torch.device("cpu") + self.flownet.to(device) + self.contextnet.to(device) + self.fusionnet.to(device) + + def load_model(self, path, rank=0): + def convert(param): + return { + k.replace("module.", ""): v + for k, v in param.items() + if "module." in k + } + + if rank == 0: + self.flownet.load_state_dict( + convert(torch.load('{}/flownet.pkl'.format(path), map_location=torch.device("cpu")))) + self.contextnet.load_state_dict( + convert(torch.load('{}/contextnet.pkl'.format(path), map_location=torch.device("cpu")))) + self.fusionnet.load_state_dict( + convert(torch.load('{}/unet.pkl'.format(path), map_location=torch.device("cpu")))) + + def save_model(self, path, rank=0): + if rank == 0: + torch.save(self.flownet.state_dict(), + '{}/flownet.pkl'.format(path)) + torch.save(self.contextnet.state_dict(), + '{}/contextnet.pkl'.format(path)) + torch.save(self.fusionnet.state_dict(), '{}/unet.pkl'.format(path)) + + def predict(self, imgs, flow, training=True, flow_gt=None): + img0 = imgs[:, :3] + img1 = imgs[:, 3:] + c0 = self.contextnet(img0, flow) + c1 = self.contextnet(img1, -flow) + flow = F.interpolate(flow, scale_factor=2.0, mode="bilinear", + align_corners=False) * 2.0 + refine_output, warped_img0, warped_img1, warped_img0_gt, warped_img1_gt = self.fusionnet( + img0, img1, flow, c0, c1, flow_gt) + res = torch.sigmoid(refine_output[:, :3]) * 2 - 1 + mask = torch.sigmoid(refine_output[:, 3:4]) + merged_img = warped_img0 * mask + warped_img1 * (1 - mask) + pred = merged_img + res + pred = torch.clamp(pred, 0, 1) + if training: + return pred, mask, merged_img, warped_img0, warped_img1, warped_img0_gt, warped_img1_gt + else: + return pred + + def inference(self, img0, img1): + imgs = torch.cat((img0, img1), 1) + flow, _ = self.flownet(imgs) + return self.predict(imgs, flow, training=False).detach() + + def update(self, imgs, gt, learning_rate=0, mul=1, training=True, flow_gt=None): + for param_group in self.optimG.param_groups: + param_group['lr'] = learning_rate + if training: + self.train() + else: + self.eval() + flow, flow_list = self.flownet(imgs) + pred, mask, merged_img, warped_img0, warped_img1, warped_img0_gt, warped_img1_gt = self.predict( + imgs, flow, flow_gt=flow_gt) + loss_ter = self.ter(pred, gt).mean() + if training: + with torch.no_grad(): + loss_flow = torch.abs(warped_img0_gt - gt).mean() + loss_mask = torch.abs( + merged_img - gt).sum(1, True).float().detach() + loss_mask = F.interpolate(loss_mask, scale_factor=0.5, mode="bilinear", + align_corners=False).detach() + flow_gt = (F.interpolate(flow_gt, scale_factor=0.5, mode="bilinear", + align_corners=False) * 0.5).detach() + loss_cons = 0 + for i in range(3): + loss_cons += self.epe(flow_list[i], flow_gt[:, :2], 1) + loss_cons += self.epe(-flow_list[i], flow_gt[:, 2:4], 1) + loss_cons = loss_cons.mean() * 0.01 + else: + loss_cons = torch.tensor([0]) + loss_flow = torch.abs(warped_img0 - gt).mean() + loss_mask = 1 + loss_l1 = (((pred - gt) ** 2 + 1e-6) ** 0.5).mean() + if training: + self.optimG.zero_grad() + loss_G = loss_l1 + loss_cons + loss_ter + loss_G.backward() + self.optimG.step() + return pred, merged_img, flow, loss_l1, loss_flow, loss_cons, loss_ter, loss_mask + + +if __name__ == '__main__': + img0 = torch.zeros(3, 3, 256, 256).float().to(device) + img1 = torch.tensor(np.random.normal( + 0, 1, (3, 3, 256, 256))).float().to(device) + imgs = torch.cat((img0, img1), 1) + model = Model() + model.eval() + print(model.inference(imgs).shape) diff --git a/gimp-plugins/RIFE/model/RIFE2F.py b/gimp-plugins/RIFE/model/RIFE2F.py new file mode 100755 index 0000000..feea1df --- /dev/null +++ b/gimp-plugins/RIFE/model/RIFE2F.py @@ -0,0 +1,250 @@ +import torch +import torch.nn as nn +import numpy as np +from torch.optim import AdamW +import torch.optim as optim +import itertools +from model.warplayer import warp +from torch.nn.parallel import DistributedDataParallel as DDP +from model.IFNet2F import * +import torch.nn.functional as F +from model.loss import * + +device = torch.device("cuda" if torch.cuda.is_available() else "cpu") + + +def conv(in_planes, out_planes, kernel_size=3, stride=1, padding=1, dilation=1): + return nn.Sequential( + nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, + padding=padding, dilation=dilation, bias=True), + nn.PReLU(out_planes) + ) + + +def deconv(in_planes, out_planes, kernel_size=4, stride=2, padding=1): + return nn.Sequential( + torch.nn.ConvTranspose2d(in_channels=in_planes, out_channels=out_planes, + kernel_size=4, stride=2, padding=1, bias=True), + nn.PReLU(out_planes) + ) + +def conv_woact(in_planes, out_planes, kernel_size=3, stride=1, padding=1, dilation=1): + return nn.Sequential( + nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, + padding=padding, dilation=dilation, bias=True), + ) + +class ResBlock(nn.Module): + def __init__(self, in_planes, out_planes, stride=2): + super(ResBlock, self).__init__() + if in_planes == out_planes and stride == 1: + self.conv0 = nn.Identity() + else: + self.conv0 = nn.Conv2d(in_planes, out_planes, + 3, stride, 1, bias=False) + self.conv1 = conv(in_planes, out_planes, 3, stride, 1) + self.conv2 = conv_woact(out_planes, out_planes, 3, 1, 1) + self.relu1 = nn.PReLU(1) + self.relu2 = nn.PReLU(out_planes) + self.fc1 = nn.Conv2d(out_planes, 16, kernel_size=1, bias=False) + self.fc2 = nn.Conv2d(16, out_planes, kernel_size=1, bias=False) + + def forward(self, x): + y = self.conv0(x) + x = self.conv1(x) + x = self.conv2(x) + w = x.mean(3, True).mean(2, True) + w = self.relu1(self.fc1(w)) + w = torch.sigmoid(self.fc2(w)) + x = self.relu2(x * w + y) + return x + +c = 16 + +class ContextNet(nn.Module): + def __init__(self): + super(ContextNet, self).__init__() + self.conv1 = ResBlock(3, c, 1) + self.conv2 = ResBlock(c, 2*c) + self.conv3 = ResBlock(2*c, 4*c) + self.conv4 = ResBlock(4*c, 8*c) + + def forward(self, x, flow): + x = self.conv1(x) + f1 = warp(x, flow) + x = self.conv2(x) + flow = F.interpolate(flow, scale_factor=0.5, mode="bilinear", + align_corners=False) * 0.5 + f2 = warp(x, flow) + x = self.conv3(x) + flow = F.interpolate(flow, scale_factor=0.5, mode="bilinear", + align_corners=False) * 0.5 + f3 = warp(x, flow) + x = self.conv4(x) + flow = F.interpolate(flow, scale_factor=0.5, mode="bilinear", + align_corners=False) * 0.5 + f4 = warp(x, flow) + return [f1, f2, f3, f4] + + +class FusionNet(nn.Module): + def __init__(self): + super(FusionNet, self).__init__() + self.down0 = ResBlock(8, 2*c, 1) + self.down1 = ResBlock(4*c, 4*c) + self.down2 = ResBlock(8*c, 8*c) + self.down3 = ResBlock(16*c, 16*c) + self.up0 = deconv(32*c, 8*c) + self.up1 = deconv(16*c, 4*c) + self.up2 = deconv(8*c, 2*c) + self.up3 = deconv(4*c, c) + self.conv = nn.Conv2d(c, 4, 3, 2, 1) + + def forward(self, img0, img1, flow, c0, c1, flow_gt): + warped_img0 = warp(img0, flow) + warped_img1 = warp(img1, -flow) + if flow_gt == None: + warped_img0_gt, warped_img1_gt = None, None + else: + warped_img0_gt = warp(img0, flow_gt[:, :2]) + warped_img1_gt = warp(img1, flow_gt[:, 2:4]) + s0 = self.down0(torch.cat((warped_img0, warped_img1, flow), 1)) + s1 = self.down1(torch.cat((s0, c0[0], c1[0]), 1)) + s2 = self.down2(torch.cat((s1, c0[1], c1[1]), 1)) + s3 = self.down3(torch.cat((s2, c0[2], c1[2]), 1)) + x = self.up0(torch.cat((s3, c0[3], c1[3]), 1)) + x = self.up1(torch.cat((x, s2), 1)) + x = self.up2(torch.cat((x, s1), 1)) + x = self.up3(torch.cat((x, s0), 1)) + x = self.conv(x) + return x, warped_img0, warped_img1, warped_img0_gt, warped_img1_gt + + +class Model: + def __init__(self, local_rank=-1): + self.flownet = IFNet() + self.contextnet = ContextNet() + self.fusionnet = FusionNet() + self.device() + self.optimG = AdamW(itertools.chain( + self.flownet.parameters(), + self.contextnet.parameters(), + self.fusionnet.parameters()), lr=1e-6, weight_decay=1e-5) + self.schedulerG = optim.lr_scheduler.CyclicLR( + self.optimG, base_lr=1e-6, max_lr=1e-3, step_size_up=8000, cycle_momentum=False) + self.epe = EPE() + self.ter = Ternary() + self.sobel = SOBEL() + if local_rank != -1: + self.flownet = DDP(self.flownet, device_ids=[ + local_rank], output_device=local_rank) + self.contextnet = DDP(self.contextnet, device_ids=[ + local_rank], output_device=local_rank) + self.fusionnet = DDP(self.fusionnet, device_ids=[ + local_rank], output_device=local_rank) + + def train(self): + self.flownet.train() + self.contextnet.train() + self.fusionnet.train() + + def eval(self): + self.flownet.eval() + self.contextnet.eval() + self.fusionnet.eval() + + def device(self): + self.flownet.to(device) + self.contextnet.to(device) + self.fusionnet.to(device) + + def load_model(self, path, rank=0): + def convert(param): + return { + k.replace("module.", ""): v + for k, v in param.items() + if "module." in k + } + if rank == 0: + self.flownet.load_state_dict( + convert(torch.load('{}/flownet.pkl'.format(path), map_location=device))) + self.contextnet.load_state_dict( + convert(torch.load('{}/contextnet.pkl'.format(path), map_location=device))) + self.fusionnet.load_state_dict( + convert(torch.load('{}/unet.pkl'.format(path), map_location=device))) + + def save_model(self, path, rank=0): + if rank == 0: + torch.save(self.flownet.state_dict(), + '{}/flownet.pkl'.format(path)) + torch.save(self.contextnet.state_dict(), + '{}/contextnet.pkl'.format(path)) + torch.save(self.fusionnet.state_dict(), '{}/unet.pkl'.format(path)) + + def predict(self, imgs, flow, training=True, flow_gt=None): + img0 = imgs[:, :3] + img1 = imgs[:, 3:] + flow = F.interpolate(flow, scale_factor=2.0, mode="bilinear", + align_corners=False) * 2.0 + c0 = self.contextnet(img0, flow) + c1 = self.contextnet(img1, -flow) + refine_output, warped_img0, warped_img1, warped_img0_gt, warped_img1_gt = self.fusionnet( + img0, img1, flow, c0, c1, flow_gt) + res = torch.sigmoid(refine_output[:, :3]) * 2 - 1 + mask = torch.sigmoid(refine_output[:, 3:4]) + merged_img = warped_img0 * mask + warped_img1 * (1 - mask) + pred = merged_img + res + pred = torch.clamp(pred, 0, 1) + if training: + return pred, mask, merged_img, warped_img0, warped_img1, warped_img0_gt, warped_img1_gt + else: + return pred + + def inference(self, img0, img1): + with torch.no_grad(): + imgs = torch.cat((img0, img1), 1) + flow, _ = self.flownet(imgs) + return self.predict(imgs, flow, training=False).detach() + + def update(self, imgs, gt, learning_rate=0, mul=1, training=True, flow_gt=None): + for param_group in self.optimG.param_groups: + param_group['lr'] = learning_rate + if training: + self.train() + else: + self.eval() + flow, flow_list = self.flownet(imgs) + pred, mask, merged_img, warped_img0, warped_img1, warped_img0_gt, warped_img1_gt = self.predict( + imgs, flow, flow_gt=flow_gt) + loss_ter = self.ter(pred, gt).mean() + if training: + with torch.no_grad(): + loss_flow = torch.abs(warped_img0_gt - gt).mean() + loss_mask = torch.abs( + merged_img - gt).sum(1, True).float().detach() + loss_cons = 0 + for i in range(3): + loss_cons += self.epe(flow_list[i], flow_gt[:, :2], 1) + loss_cons += self.epe(-flow_list[i], flow_gt[:, 2:4], 1) + loss_cons = loss_cons.mean() * 0.01 + else: + loss_cons = torch.tensor([0]) + loss_flow = torch.abs(warped_img0 - gt).mean() + loss_mask = 1 + loss_l1 = (((pred - gt) ** 2 + 1e-6) ** 0.5).mean() + if training: + self.optimG.zero_grad() + loss_G = loss_l1 + loss_cons + loss_ter + loss_G.backward() + self.optimG.step() + return pred, merged_img, flow, loss_l1, loss_flow, loss_cons, loss_ter, loss_mask + + +if __name__ == '__main__': + img0 = torch.zeros(3, 3, 256, 256).float().to(device) + img1 = torch.tensor(np.random.normal( + 0, 1, (3, 3, 256, 256))).float().to(device) + imgs = torch.cat((img0, img1), 1) + model = Model() + model.eval() + print(model.inference(imgs).shape) diff --git a/gimp-plugins/RIFE/model/__init__.py b/gimp-plugins/RIFE/model/__init__.py new file mode 100755 index 0000000..e69de29 diff --git a/gimp-plugins/RIFE/model/loss.py b/gimp-plugins/RIFE/model/loss.py new file mode 100755 index 0000000..a809d97 --- /dev/null +++ b/gimp-plugins/RIFE/model/loss.py @@ -0,0 +1,90 @@ +import torch +import numpy as np +import torch.nn as nn +import torch.nn.functional as F + +# device = torch.device("cuda" if torch.cuda.is_available() else "cpu") + + +class EPE(nn.Module): + def __init__(self): + super(EPE, self).__init__() + + def forward(self, flow, gt, loss_mask): + loss_map = (flow - gt.detach()) ** 2 + loss_map = (loss_map.sum(1, True) + 1e-6) ** 0.5 + return (loss_map * loss_mask) + + +class Ternary(nn.Module): + def __init__(self, cFlag): + super(Ternary, self).__init__() + patch_size = 7 + out_channels = patch_size * patch_size + self.w = np.eye(out_channels).reshape( + (patch_size, patch_size, 1, out_channels)) + self.w = np.transpose(self.w, (3, 2, 0, 1)) + self.device = torch.device("cuda" if torch.cuda.is_available() and not cFlag else "cpu") + self.w = torch.tensor(self.w).float().to(self.device) + + def transform(self, img): + patches = F.conv2d(img, self.w, padding=3, bias=None) + transf = patches - img + transf_norm = transf / torch.sqrt(0.81 + transf**2) + return transf_norm + + def rgb2gray(self, rgb): + r, g, b = rgb[:, 0:1, :, :], rgb[:, 1:2, :, :], rgb[:, 2:3, :, :] + gray = 0.2989 * r + 0.5870 * g + 0.1140 * b + return gray + + def hamming(self, t1, t2): + dist = (t1 - t2) ** 2 + dist_norm = torch.mean(dist / (0.1 + dist), 1, True) + return dist_norm + + def valid_mask(self, t, padding): + n, _, h, w = t.size() + inner = torch.ones(n, 1, h - 2 * padding, w - 2 * padding).type_as(t) + mask = F.pad(inner, [padding] * 4) + return mask + + def forward(self, img0, img1): + img0 = self.transform(self.rgb2gray(img0)) + img1 = self.transform(self.rgb2gray(img1)) + return self.hamming(img0, img1) * self.valid_mask(img0, 1) + + +class SOBEL(nn.Module): + def __init__(self, cFlag): + super(SOBEL, self).__init__() + self.kernelX = torch.tensor([ + [1, 0, -1], + [2, 0, -2], + [1, 0, -1], + ]).float() + self.kernelY = self.kernelX.clone().T + self.device = torch.device("cuda" if torch.cuda.is_available() and not cFlag else "cpu") + self.kernelX = self.kernelX.unsqueeze(0).unsqueeze(0).to(self.device) + self.kernelY = self.kernelY.unsqueeze(0).unsqueeze(0).to(self.device) + + def forward(self, pred, gt): + N, C, H, W = pred.shape[0], pred.shape[1], pred.shape[2], pred.shape[3] + img_stack = torch.cat( + [pred.reshape(N*C, 1, H, W), gt.reshape(N*C, 1, H, W)], 0) + sobel_stack_x = F.conv2d(img_stack, self.kernelX, padding=1) + sobel_stack_y = F.conv2d(img_stack, self.kernelY, padding=1) + pred_X, gt_X = sobel_stack_x[:N*C], sobel_stack_x[N*C:] + pred_Y, gt_Y = sobel_stack_y[:N*C], sobel_stack_y[N*C:] + + L1X, L1Y = torch.abs(pred_X-gt_X), torch.abs(pred_Y-gt_Y) + loss = (L1X+L1Y) + return loss + + +if __name__ == '__main__': + img0 = torch.zeros(3, 3, 256, 256).float().to(device) + img1 = torch.tensor(np.random.normal( + 0, 1, (3, 3, 256, 256))).float().to(device) + ternary_loss = Ternary() + print(ternary_loss(img0, img1).shape) diff --git a/gimp-plugins/RIFE/model/warplayer.py b/gimp-plugins/RIFE/model/warplayer.py new file mode 100755 index 0000000..dfc7e41 --- /dev/null +++ b/gimp-plugins/RIFE/model/warplayer.py @@ -0,0 +1,23 @@ +import torch +import torch.nn as nn + +backwarp_tenGrid = {} + + +def warp(tenInput, tenFlow, cFlag): + device = torch.device("cuda" if torch.cuda.is_available() and not cFlag else "cpu") + k = (str(tenFlow.device), str(tenFlow.size())) + if k not in backwarp_tenGrid: + tenHorizontal = torch.linspace(-1.0, 1.0, tenFlow.shape[3]).view( + 1, 1, 1, tenFlow.shape[3]).expand(tenFlow.shape[0], -1, tenFlow.shape[2], -1) + tenVertical = torch.linspace(-1.0, 1.0, tenFlow.shape[2]).view( + 1, 1, tenFlow.shape[2], 1).expand(tenFlow.shape[0], -1, -1, tenFlow.shape[3]) + backwarp_tenGrid[k] = torch.cat( + [tenHorizontal, tenVertical], 1).to(device) + + tenFlow = torch.cat([tenFlow[:, 0:1, :, :] / ((tenInput.shape[3] - 1.0) / 2.0), + tenFlow[:, 1:2, :, :] / ((tenInput.shape[2] - 1.0) / 2.0)], 1) + + g = (backwarp_tenGrid[k] + tenFlow).permute(0, 2, 3, 1) + return torch.nn.functional.grid_sample(input=tenInput, grid=torch.clamp(g, -1, 1), mode='bilinear', + padding_mode='zeros', align_corners=True) diff --git a/gimp-plugins/interpolateframes.py b/gimp-plugins/interpolateframes.py new file mode 100755 index 0000000..96cc8f0 --- /dev/null +++ b/gimp-plugins/interpolateframes.py @@ -0,0 +1,126 @@ +import os + +baseLoc = os.path.dirname(os.path.realpath(__file__)) + '/' +savePath = '/'.join(baseLoc.split('/')[:-2]) + '/output/interpolateframes' +from gimpfu import * +import sys + +sys.path.extend([baseLoc + 'gimpenv/lib/python2.7', baseLoc + 'gimpenv/lib/python2.7/site-packages', + baseLoc + 'gimpenv/lib/python2.7/site-packages/setuptools', baseLoc + 'RIFE']) + +import cv2 +import torch +from torch.nn import functional as F +from model import RIFE +import numpy as np + + + +def channelData(layer): # convert gimp image to numpy + region = layer.get_pixel_rgn(0, 0, layer.width, layer.height) + pixChars = region[:, :] # Take whole layer + bpp = region.bpp + # return np.frombuffer(pixChars,dtype=np.uint8).reshape(len(pixChars)/bpp,bpp) + return np.frombuffer(pixChars, dtype=np.uint8).reshape(layer.height, layer.width, bpp) + + +def createResultLayer(image, name, result): + rlBytes = np.uint8(result).tobytes(); + rl = gimp.Layer(image, name, image.width, image.height, 0, 100, NORMAL_MODE) + region = rl.get_pixel_rgn(0, 0, rl.width, rl.height, True) + region[:, :] = rlBytes + image.add_layer(rl, 0) + gimp.displays_flush() + + +def getinter(img_s, img_e, c_flag, string_path): + exp = 4 + out_path = string_path + + model = RIFE.Model(c_flag) + model.load_model(baseLoc + 'weights' + '/interpolateframes') + model.eval() + model.device(c_flag) + + img0 = img_s + img1 = img_e + + img0 = (torch.tensor(img0.transpose(2, 0, 1)) / 255.).unsqueeze(0) + img1 = (torch.tensor(img1.transpose(2, 0, 1))/ 255.).unsqueeze(0) + if torch.cuda.is_available() and not c_flag: + device = torch.device("cuda") + else: + device = torch.device("cpu") + + img0 = img0.to(device) + img1 = img1.to(device) + + n, c, h, w = img0.shape + ph = ((h - 1) // 32 + 1) * 32 + pw = ((w - 1) // 32 + 1) * 32 + padding = (0, pw - w, 0, ph - h) + img0 = F.pad(img0, padding) + img1 = F.pad(img1, padding) + + img_list = [img0, img1] + idx=0 + t=exp * (len(img_list) - 1) + for i in range(exp): + tmp = [] + for j in range(len(img_list) - 1): + mid = model.inference(img_list[j], img_list[j + 1]) + tmp.append(img_list[j]) + tmp.append(mid) + idx=idx+1 + gimp.progress_update(float(idx)/float(t)) + gimp.displays_flush() + tmp.append(img1) + img_list = tmp + + if not os.path.exists(out_path): + os.makedirs(out_path) + for i in range(len(img_list)): + cv2.imwrite(out_path + '/img{}.png'.format(i), + (img_list[i][0] * 255).byte().cpu().numpy().transpose(1, 2, 0)[:h, :w, ::-1]) + + + + +def interpolateframes(imggimp, curlayer, string_path, layer_s, layer_e, c_flag): + if torch.cuda.is_available() and not c_flag: + gimp.progress_init("(Using GPU) Running slomo and saving frames in "+string_path) + # device = torch.device("cuda") + else: + gimp.progress_init("(Using CPU) Running slomo and saving frames in "+string_path) + # device = torch.device("cpu") + + layer_1 = channelData(layer_s) + layer_2 = channelData(layer_e) + if layer_1.shape[2] == 4: # get rid of alpha channel + layer_1 = layer_1[:, :, 0:3] + if layer_2.shape[2] == 4: # get rid of alpha channel + layer_2 = layer_2[:, :, 0:3] + getinter(layer_1, layer_2, c_flag, string_path) + # pdb.gimp_message("Saved") + + +register( + "interpolate-frames", + "interpolate-frames", + "Running slomo...", + "Kritik Soman", + "Your", + "2020", + "interpolate-frames...", + "*", # Alternately use RGB, RGB*, GRAY*, INDEXED etc. + [(PF_IMAGE, "image", "Input image", None), + (PF_DRAWABLE, "drawable", "Input drawable", None), + (PF_STRING, "string", "Output Location", savePath), + (PF_LAYER, "drawinglayer", "Start Frame:", None), + (PF_LAYER, "drawinglayer", "End Frame:", None), + (PF_BOOL, "fcpu", "Force CPU", False) + ], + [], + interpolateframes, menu="/Layer/GIML-ML") + +main() diff --git a/gimp-plugins/syncWeights.py b/gimp-plugins/syncWeights.py index ffac09a..122d8de 100755 --- a/gimp-plugins/syncWeights.py +++ b/gimp-plugins/syncWeights.py @@ -215,4 +215,31 @@ def sync(path,flag): gimp.progress_init("Downloading " + model +"(~" + str(fileSize) + "MB)...") download_file_from_google_drive(file_id, destination,fileSize) + #interpolateframes + model = 'interpolateframes' + file_id = '1bHmO9-_ENTYoN1-BNwSk3nLN9-NDUnRg' + fileSize = 1.6 #in MB + mFName = 'contextnet.pkl' + if not os.path.isdir(path + '/' + model): + os.mkdir(path + '/' + model) + destination = path + '/' + model + '/' + mFName + if not os.path.isfile(destination): + gimp.progress_init("Downloading " + model +"(~" + str(fileSize) + "MB)...") + download_file_from_google_drive(file_id, destination,fileSize) + file_id = '1cQvDPBKsz3TAi0Q5bJXsu6A-Z7lpk_cE' + fileSize = 25.4 #in MB + mFName = 'flownet.pkl' + destination = path + '/' + model + '/' + mFName + if not os.path.isfile(destination): + gimp.progress_init("Downloading " + model +"(~" + str(fileSize) + "MB)...") + download_file_from_google_drive(file_id, destination,fileSize) + file_id = '1mlA8VtxIcvJfz51OsQMvWX24oqxZ429r' + fileSize = 15 #in MB + mFName = 'unet.pkl' + destination = path + '/' + model + '/' + mFName + if not os.path.isfile(destination): + gimp.progress_init("Downloading " + model +"(~" + str(fileSize) + "MB)...") + download_file_from_google_drive(file_id, destination,fileSize) + +