生成的头像部分和身体部分错位了

[Mou-Mou-L]

不知道怎么解决
parser = ArgumentParser()
parser.add_argument("--driven_audio", default='examples/driven_audio/2.wav',
help="path to driven audio")
parser.add_argument("--source_video", default='examples/driven_video/3.mp4',
help="path to source video")
parser.add_argument("--checkpoint_dir", default='checkpoints', help="path to output")
parser.add_argument("--result_dir", default='./results', help="path to output")
parser.add_argument("--batch_size", type=int, default=8, help="the batch size of facerender")
parser.add_argument("--enhancer", type=str, default='none', help="enhaner region:[none,lip,face]
none:do not enhance;
lip:only enhance lip region
face: enhance (skin nose eye brow lip) region")
parser.add_argument("--cpu", dest="cpu", action="store_true")

parser.add_argument("--use_DAIN", dest="use_DAIN", action="store_true",
                    help="Depth-Aware Video Frame Interpolation")
parser.add_argument('--DAIN_weight', type=str, default='./checkpoints/DAIN_weight',
                    help='Path to model weight')
parser.add_argument('--dian_output', type=str, default='dian_output', help='output dir')
parser.add_argument('--time_step', type=float, default=0.5, help='choose the time steps')
parser.add_argument('--remove_duplicates', action='store_true', default=False,
                    help='whether to remove duplicated frames')

args = parser.parse_args()
if torch.cuda.is_available() and not args.cpu:
    args.device = "cuda"
else:
    args.device = "cpu"

args.use_DAIN = True
main(args)

参数如上
python=3.10 显卡如下
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Fri_Jun_14_16:44:19_Pacific_Daylight_Time_2024
Cuda compilation tools, release 12.6, V12.6.20
Build cuda_12.6.r12.6/compiler.34431801_0