render.py

#
# 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
from scene import Scene
import os
from tqdm import tqdm
from os import makedirs
from gaussian_renderer import render
import torchvision
from utils.general_utils import safe_state
from argparse import ArgumentParser
from arguments import ModelParams, PipelineParams, get_combined_args
from gaussian_renderer import GaussianModel
import numpy as np
import cv2
from scene.cameras import Camera
from utils.graphics_utils import getWorld2View2, getProjectionMatrix
try:
    from diff_gaussian_rasterization import SparseGaussianAdam
    SPARSE_ADAM_AVAILABLE = True
except:
    SPARSE_ADAM_AVAILABLE = False


def render_set(model_path, name, iteration, views, gaussians, pipeline, background, train_test_exp, separate_sh):
    render_path = os.path.join(model_path, name, "ours_{}".format(iteration), "renders")
    gts_path = os.path.join(model_path, name, "ours_{}".format(iteration), "gt")

    makedirs(render_path, exist_ok=True)
    makedirs(gts_path, exist_ok=True)

    for idx, view in enumerate(tqdm(views, desc="Rendering progress")):
        rendering = render(view, gaussians, pipeline, background, use_trained_exp=train_test_exp, separate_sh=separate_sh)["render"]
        gt = view.original_image[0:3, :, :]

        if args.train_test_exp:
            rendering = rendering[..., rendering.shape[-1] // 2:]
            gt = gt[..., gt.shape[-1] // 2:]

        torchvision.utils.save_image(rendering, os.path.join(render_path, '{0:05d}'.format(idx) + ".png"))
        torchvision.utils.save_image(gt, os.path.join(gts_path, '{0:05d}'.format(idx) + ".png"))

def create_rotation_views(camera_pos, avg_up, num_frames=120):
    """固定位置，围绕up轴360度旋转朝向"""
    angles = np.linspace(0, 2 * np.pi, num_frames, endpoint=False)
    
    # 创建一个垂直于up向量的基准方向
    # 找一个不平行于up的向量
    if abs(np.dot(avg_up, [1, 0, 0])) < 0.9:
        base_dir = np.array([1, 0, 0])
    else:
        base_dir = np.array([0, 1, 0])
    
    # 计算垂直于up的两个正交向量
    right = np.cross(avg_up, base_dir)
    right = right / np.linalg.norm(right)
    forward_base = np.cross(right, avg_up)
    forward_base = forward_base / np.linalg.norm(forward_base)
    
    look_at_directions = []
    for angle in angles:
        # 围绕up轴旋转
        cos_a, sin_a = np.cos(angle), np.sin(angle)
        forward = cos_a * forward_base + sin_a * right
        look_at_directions.append(forward)
    
    return look_at_directions

def render_rotation_video(model_path, iteration, gaussians, pipeline, background, camera_pos, avg_up, train_cameras, num_frames=480, fps=30):
    """生成旋转视频"""
    video_path = os.path.join(model_path, "rotation_video")
    makedirs(video_path, exist_ok=True)
    
    # 创建旋转朝向
    look_at_directions = create_rotation_views(camera_pos, avg_up, num_frames)
    
    # 从训练相机获取真实参数
    ref_cam = train_cameras[0]
    width = ref_cam.image_width
    height_cam = ref_cam.image_height
    fovx = ref_cam.FoVx
    fovy = ref_cam.FoVy
    znear = 0.01
    zfar = 100.0
    
    frames = []
    
    for i, look_direction in enumerate(tqdm(look_at_directions, desc="Rendering rotation video")):
        # 相机位置固定，只改变朝向
        pos = camera_pos
        forward = look_direction
        forward = forward / np.linalg.norm(forward)
        
        # 使用从训练相机提取的平均up向量
        up = avg_up
        right = np.cross(forward, up)
        right = right / np.linalg.norm(right)
        up = np.cross(right, forward)  # 重新计算up保证正交
        
        # 使用getWorld2View2构建变换矩阵
        R = np.array([right, up, -forward]).T
        t = -R @ pos
        
        world_view_transform = torch.tensor(getWorld2View2(R, t)).transpose(0, 1).cuda()
        projection_matrix = getProjectionMatrix(znear, zfar, fovx, fovy).transpose(0, 1).cuda()
        full_proj_transform = world_view_transform.unsqueeze(0).bmm(projection_matrix.unsqueeze(0)).squeeze(0)
        camera_center = world_view_transform.inverse()[3, :3]
        
        # 创建虚拟相机
        class VirtualCamera:
            def __init__(self):
                self.world_view_transform = world_view_transform
                self.projection_matrix = projection_matrix
                self.full_proj_transform = full_proj_transform
                self.camera_center = camera_center
                self.image_width = width
                self.image_height = height_cam
                self.FoVx = fovx
                self.FoVy = fovy
        
        virtual_cam = VirtualCamera()
        
        # 渲染
        rendering = render(virtual_cam, gaussians, pipeline, background)["render"]
        
        # 转换为numpy数组并保存帧
        frame = rendering.detach().cpu().numpy().transpose(1, 2, 0)
        frame = (frame * 255).astype(np.uint8)
        frames.append(frame)
        
        # 保存单帧图像
        torchvision.utils.save_image(rendering, os.path.join(video_path, f'{i:05d}.png'))
    
    # 生成视频
    video_file = os.path.join(video_path, 'circular_video.mp4')
    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
    out = cv2.VideoWriter(video_file, fourcc, fps, (frames[0].shape[1], frames[0].shape[0]))
    
    for frame in frames:
        frame_bgr = cv2.cvtColor(frame, cv2.COLOR_RGB2BGR)
        out.write(frame_bgr)
    
    out.release()
    print(f"旋转视频已保存到: {video_file}")

def render_sets(dataset : ModelParams, iteration : int, pipeline : PipelineParams, skip_train : bool, skip_test : bool, separate_sh: bool, render_video: bool = False):
    with torch.no_grad():
        gaussians = GaussianModel(dataset.sh_degree)
        scene = Scene(dataset, gaussians, load_iteration=iteration, shuffle=False)

        bg_color = [1,1,1] if dataset.white_background else [0, 0, 0]
        background = torch.tensor(bg_color, dtype=torch.float32, device="cuda")

        if not skip_train:
             render_set(dataset.model_path, "train", scene.loaded_iter, scene.getTrainCameras(), gaussians, pipeline, background, dataset.train_test_exp, separate_sh)

        if not skip_test:
             render_set(dataset.model_path, "test", scene.loaded_iter, scene.getTestCameras(), gaussians, pipeline, background, dataset.train_test_exp, separate_sh)
        
        if render_video:
            # 计算场景中心和合适的相机参数
            train_cameras = scene.getTrainCameras()
            if train_cameras:
                # 从训练相机估算场景中心
                positions = [cam.camera_center.cpu().numpy() for cam in train_cameras]
                center = np.mean(positions, axis=0)
                
                # 使用第104个相机（索引103）的Y轴作为up向量
                if len(train_cameras) > 110:
                    ref_cam_for_up = train_cameras[103]
                    R = ref_cam_for_up.world_view_transform[:3, :3].cpu().numpy()
                    avg_up = R[0, :]  # 第104个相机的X轴（向上方向）
                    avg_up = avg_up / np.linalg.norm(avg_up)  # 归一化
                else:
                    # 如果相机数量不够，使用第一个相机的up向量
                    R = train_cameras[0].world_view_transform[:3, :3].cpu().numpy()
                    avg_up = R[0, :]
                    avg_up = avg_up / np.linalg.norm(avg_up)
                
                render_rotation_video(dataset.model_path, scene.loaded_iter, gaussians, pipeline, background, 
                                    center, avg_up, train_cameras)

if __name__ == "__main__":
    # Set up command line argument parser
    parser = ArgumentParser(description="Testing script parameters")
    model = ModelParams(parser, sentinel=True)
    pipeline = PipelineParams(parser)
    parser.add_argument("--iteration", default=-1, type=int)
    parser.add_argument("--skip_train", action="store_true")
    parser.add_argument("--skip_test", action="store_true")
    parser.add_argument("--quiet", action="store_true")
    parser.add_argument("--render_video", action="store_true", help="生成环形视频")
    args = get_combined_args(parser)
    print("Rendering " + args.model_path)

    # Initialize system state (RNG)
    safe_state(args.quiet)

    render_sets(model.extract(args), args.iteration, pipeline.extract(args), args.skip_train, args.skip_test, SPARSE_ADAM_AVAILABLE, args.render_video)