# # Copyright (C) 2023, Inria # GRAPHDECO research group, https://team.inria.fr/graphdeco # All rights reserved. # # This software is free for non-commercial, research and evaluation use # under the terms of the LICENSE.md file. # # For inquiries contact george.drettakis@inria.fr # import torch import torch.nn.functional as F from torch.autograd import Variable from math import exp def l1_loss(network_output, gt): return torch.abs((network_output - gt)).mean() def l2_loss(network_output, gt): return ((network_output - gt) ** 2).mean() def gaussian(window_size, sigma): gauss = torch.Tensor([exp(-(x - window_size // 2) ** 2 / float(2 * sigma ** 2)) for x in range(window_size)]) return gauss / gauss.sum() def create_window(window_size, channel): _1D_window = gaussian(window_size, 1.5).unsqueeze(1) _2D_window = _1D_window.mm(_1D_window.t()).float().unsqueeze(0).unsqueeze(0) window = Variable(_2D_window.expand(channel, 1, window_size, window_size).contiguous()) return window def ssim(img1, img2, window_size=11, size_average=True): channel = img1.size(-3) window = create_window(window_size, channel) if img1.is_cuda: window = window.cuda(img1.get_device()) window = window.type_as(img1) return _ssim(img1, img2, window, window_size, channel, size_average) def _ssim(img1, img2, window, window_size, channel, size_average=True): mu1 = F.conv2d(img1, window, padding=window_size // 2, groups=channel) mu2 = F.conv2d(img2, window, padding=window_size // 2, groups=channel) mu1_sq = mu1.pow(2) mu2_sq = mu2.pow(2) mu1_mu2 = mu1 * mu2 sigma1_sq = F.conv2d(img1 * img1, window, padding=window_size // 2, groups=channel) - mu1_sq sigma2_sq = F.conv2d(img2 * img2, window, padding=window_size // 2, groups=channel) - mu2_sq sigma12 = F.conv2d(img1 * img2, window, padding=window_size // 2, groups=channel) - mu1_mu2 C1 = 0.01 ** 2 C2 = 0.03 ** 2 ssim_map = ((2 * mu1_mu2 + C1) * (2 * sigma12 + C2)) / ((mu1_sq + mu2_sq + C1) * (sigma1_sq + sigma2_sq + C2)) if size_average: return ssim_map.mean() else: return ssim_map.mean(1).mean(1).mean(1) import torch import torch.nn as nn from taming.modules.losses.vqperceptual import * # TODO: taming dependency yes/no? class LPIPSWithDiscriminator(nn.Module): def __init__(self, disc_start, logvar_init=0.0, kl_weight=1.0, pixelloss_weight=1.0, disc_num_layers=3, disc_in_channels=3, disc_factor=1.0, disc_weight=1.0, perceptual_weight=1.0, use_actnorm=False, disc_conditional=False, disc_loss="hinge"): super().__init__() assert disc_loss in ["hinge", "vanilla"] self.kl_weight = kl_weight self.pixel_weight = pixelloss_weight self.perceptual_loss = LPIPS().eval() self.perceptual_weight = perceptual_weight # output log variance self.logvar = nn.Parameter(torch.ones(size=()) * logvar_init) self.discriminator = NLayerDiscriminator(input_nc=disc_in_channels, n_layers=disc_num_layers, use_actnorm=use_actnorm ).apply(weights_init) self.discriminator_iter_start = disc_start self.disc_loss = hinge_d_loss if disc_loss == "hinge" else vanilla_d_loss self.disc_factor = disc_factor self.discriminator_weight = disc_weight self.disc_conditional = disc_conditional def calculate_adaptive_weight(self, nll_loss, g_loss, last_layer=None): if last_layer is not None: nll_grads = torch.autograd.grad(nll_loss, last_layer, retain_graph=True)[0] g_grads = torch.autograd.grad(g_loss, last_layer, retain_graph=True)[0] else: nll_grads = torch.autograd.grad(nll_loss, self.last_layer[0], retain_graph=True)[0] g_grads = torch.autograd.grad(g_loss, self.last_layer[0], retain_graph=True)[0] d_weight = torch.norm(nll_grads) / (torch.norm(g_grads) + 1e-4) d_weight = torch.clamp(d_weight, 0.0, 1e4).detach() d_weight = d_weight * self.discriminator_weight return d_weight def forward(self, inputs, reconstructions, optimizer_idx, global_step, last_layer=None, cond=None, split="train"): rec_loss = torch.abs(inputs.contiguous() - reconstructions.contiguous()) if self.perceptual_weight > 0: p_loss = self.perceptual_loss(inputs.contiguous(), reconstructions.contiguous()) rec_loss = rec_loss + self.perceptual_weight * p_loss # nll_loss = rec_loss / torch.exp(self.logvar) + self.logvar # now the GAN part if optimizer_idx == 0: # generator update logits_fake = self.discriminator(reconstructions.contiguous()) # g_loss = -torch.mean(logits_fake) g_loss = F.relu(1 - logits_fake).mean() # if self.disc_factor > 0.0: # try: # d_weight = self.calculate_adaptive_weight(nll_loss, g_loss, last_layer=last_layer) # except RuntimeError: # assert not self.training # d_weight = torch.tensor(0.0) # else: # d_weight = torch.tensor(0.0) # disc_factor = adopt_weight(self.disc_factor, global_step, threshold=self.discriminator_iter_start) # loss = d_weight * disc_factor * g_loss # return loss, log return g_loss if optimizer_idx == 1: # second pass for discriminator update logits_real = self.discriminator(inputs.contiguous().detach()) logits_fake = self.discriminator(reconstructions.contiguous().detach()) # disc_factor = adopt_weight(self.disc_factor, global_step, threshold=self.discriminator_iter_start) # d_loss = disc_factor * self.disc_loss(logits_real, logits_fake) # log = {"{}/disc_loss".format(split): d_loss.clone().detach().mean(), # "{}/logits_real".format(split): logits_real.detach().mean(), # "{}/logits_fake".format(split): logits_fake.detach().mean() # } # return d_loss, log d_loss = self.disc_loss(logits_real, logits_fake) return d_loss import torch from chamfer_distance import ChamferDistance # 初始化 Chamfer Distance 模块 chamfer_dist_module = ChamferDistance() def calculate_chamfer_loss(pred, gt): """ 计算 Chamfer Distance 损失 Args: pred (torch.Tensor): 预测点云，维度为 (batch_size, num_points, 3) gt (torch.Tensor): 真实点云，维度为 (batch_size, num_points, 3) chamfer_dist_module (ChamferDistance): 预先初始化的 Chamfer Distance 模块 Returns: torch.Tensor: Chamfer Distance 损失 """ # 计算 Chamfer Distance dist1, dist2, idx1, idx2 = chamfer_dist_module(pred, gt) loss = (torch.mean(dist1) + torch.mean(dist2)) / 2 return loss if __name__ == "__main__": discriminator = LPIPSWithDiscriminator(disc_start=0, disc_weight=0.5)