Function render_sets

render.py:189–248

(
    dataset: ModelParams,
    iteration: int,
    pipeline: PipelineParams,
    skip_train: bool,
    skip_test: bool,
    args,
)

def render_sets(
    dataset: ModelParams,
    iteration: int,
    pipeline: PipelineParams,
    skip_train: bool,
    skip_test: bool,
    args,
):
    with torch.no_grad():
        gaussians = GaussianModel(dataset.sh_degree)
        scene = Scene(dataset, gaussians, load_iteration=iteration, opt=args, 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:
    if not skip_train and not args.infer_video and not dataset.eval:        
        optimized_pose = np.load(Path(args.model_path) / 'pose' / f'ours_{iteration}' / 'pose_optimized.npy')
        viewpoint_stack = loadCameras(optimized_pose, scene.getTrainCameras())
        render_set(
            dataset.model_path,
            "train",
            scene.loaded_iter,
            viewpoint_stack,
            gaussians,
            pipeline,
            background,
        )

    else:
        start_time = time()
        if not skip_test:
            render_set_optimize(
                dataset.model_path,
                "test",
                scene.loaded_iter,
                scene.getTestCameras(),
                gaussians,
                pipeline,
                background,
            )
        end_time = time()
        save_time(dataset.model_path, '[4] render', end_time - start_time)

    if args.infer_video and not dataset.eval:
        save_interpolate_pose(Path(args.model_path), iteration, args.n_views)
        interp_pose = np.load(Path(args.model_path) / 'pose' / f'ours_{iteration}' / 'pose_interpolated.npy')
        viewpoint_stack = loadCameras(interp_pose, scene.getTrainCameras())
        render_set(
            dataset.model_path,
            "interp",
            scene.loaded_iter,
            viewpoint_stack,
            gaussians,
            pipeline,
            background,
        )
        image_folder = os.path.join(dataset.model_path, f'interp/ours_{iteration}/renders')