人脸替换工具升级到最新4.8了。并回复最近小伙伴留言的一些疑问

1894.png


刚才才看到这个人脸替换工具升级到了2.3,本来想今天已经更新了那个AI图片分享就算了,但是这个工具最近很受小伙伴们的欢迎,所以临时补充一下更新吧。让大伙用上最新的版本。版本更新记录如下

V2.3版本:2023-9-20

1、更新到最新rope,修复bug,提升性能,加入一些中文字符的警告;

2、rope汉化完整;

3、人脸高清修复支持codeformer;

4、主界面勾选fp16,则换脸也会调用fp16模型,显存占用小,速度提升。

V2.2版本:2023-9-6

1、加入rope,原汁原味,仅作汉化,软件主界面下方按钮点击即可打开。


V2.1版本:2023-9-4

1、加入背景高清修复功能,对画面全局进行高清放大,推荐显卡,CPU就不要跑了,非常慢;

2、加入单人脸替换模式,对于参考人脸相似度检测不准容易出现抖动问题,单人脸替换和多人脸替换则没有相似度、年龄性别类的检测;

3、支持打开官方原汁原味的webui(facefusion)界面。


V2.0版本:2023-8-18

1、与官方next分支保持一致;

2、支持替换指定人脸,也就是参考人脸模式;

3、视频质量现在改为数字越大效果越好;

4、支持打开官方原汁原味的webui界面;

5、支持刷新电脑的摄像头参数。


V1.8版本:2023-7-30

1、支持直播换脸,选择人脸图片,点击开始直播即可,人脸替换模式速度不错,加高清修复卡顿;当选择目标视频时,实时替换视频的人脸并显示在直播窗口里;当不选择目标视频时,实时替换摄像头的人脸;

2、支持最高32线程设置;

3、记住上次选择的人脸图片或者视频目录;

4、修复特殊字符警告的bug。


V1.7版本:2023-7-28

1、生成速度优化,有所提升;

2、高清修复效果小幅优化;

3、支持视频生成的过程帧的图片格式选择和图片质量选择,临时帧默认配置下图片很大;

4、加入一些路径特殊字符的警告。


V1.6版本:2023-7-9

1、加入人脸替换+高清修复自动化处理的功能;

2、修复1.5版本批量图片替换存在覆盖的情况;

3、软件 每次 切换 处理策略,还是要重启软件,否则不生效,这个bug还没有解决。


V1.5版本:2023-7-8

1、支持批量选择源图片和目标视频/图片功能‘

2、支持生成视频后不删除帧;

3、添加快速打开输出目录按钮。


V1.4版本:2023-6-25

1、支持保持目标视频源帧率导出;

2、支持线程数设定,小的线程数显存占用小一些,但速度慢一些;

3、支持软件每次打开加载上次的训练参数。


V1.3版本:2023-6-24

1、支持多张人脸替换成同一个人脸,勾选多人脸模式即可。


V1.2版本:2023-6-24

1、修复高清修复不能使用,还有第一次报错的问题;

2、参数问号解释。


V1.1版本:2023-6-23

1、修复图片图片bug。


V1.0版本:2023-6-23

1、支持图片或者视频人脸替换;  

2、支持视频人脸高清修复

3、支持CPU或者显卡运行软件。

1、支持直播换脸,选择人脸图片,点击开始直播即可,人脸替换模式速度不错,加高清修复卡顿;当选择目标视频时,实时替换视频的人脸并显示在直播窗口里;当不选择目标视频时,实时替换摄像头的人脸;

2、支持最高32线程设置;

3、记住上次选择的人脸图片或者视频目录;

4、修复特殊字符警告的bug。

这是最新的更新记录,升级和完整包我都已经准备好了。

这里这个线程 我说明一下,不是你把线程调的越高越好,比如你的cpu本身就只有12线程,你就算调的再高也没用,而且还很容易报错。

所以大伙还是要量力而行。

下面着重回复一下最近这段时间以来大伙留言的各种疑问。

第一个 为什么生成结束后,视频没有换脸成功 还是原来的样子

这个原因有2个,1 就是你的人脸素材 不够清晰和正面。比如额头最好不要有刘海之类挡住。2就是目标视频里面人物的脸变动太大。导致换脸视频,最好是选择人物头部运行幅度比较小的,人物距离镜头稍微远点比较好。

第二个 软件无法运行,提示缺少什么DLL 之类

这里面主要原因是电脑缺少运行库

很多小伙伴 不知道运行库是什么,但是如果你经常玩游戏的话你就肯定知道这个东西,如果没有安装这个东西,就会发现游戏经常运行不了。

可以去我的网站搜索下运行库 如图所示。

528.jpg

安装好这个后,最好是重启下电脑。就可以了

第三个 点击生成视频后,那个黑黑的屏幕那里提示各种错误

这里的主要原因是 你的显卡是A卡,这个工具是不支持A卡的。所以生成设备调用那里要选CPU,而不是显卡。这样就可以避免很大一部分错误的产生了

第四个 生成的视频很模糊

模糊的产生有2个,1 本身你的视频就不高清,关于如何把模糊的视频转换成很高清的视频,可以看看这个

操作简单 最佳自动视频增强器 HitPaw Video Enhancer  可转换为动漫效果视频

先把自己的视频搞得高清一点,然后再来生成就会效果好很多。2 生成视频质量参数那里 软件本身默认是35,你可以适当调调,至于具体是多少,这个看个人视频质量来了,我一般设置是30。至于你们是不是也适合30 这个参数,你们就得慢慢试试了。

第五个  不支持win7 

以上就是最近这段时间大伙最关心的问题了。如果还有什么疑问请留言告知,另外基础的电脑知识就不要问了。比如某个小伙伴老问我怎么解压,这样的问题我实在是不知道该怎么回答了,如果这点问题都不肯自己去学习解决,那就真没有必要为难自己了。有这时间去玩几把王者难道不想吗?

以上文字全部都是手打精华,千万不要吝啬你们的赞啊和分享 



关于这个工具的介绍 大伙可以看看我前面的介绍

开源 AI人脸替换工具离线版V1.4一张图实现视频或者图片换脸,完全免费!附演示视频

上次分享的AI视频人脸替换工具 又又更新了,这次最新版是1.5 离线 免费 免安装 一键操作

免费 开源 视频图片AI人工智能人脸替换工具离线版V1.6

现在是升级到1.8了,再次感谢作者提供的这个工具,让我们很方便操作视频或者人脸替换了。

下载地址:

https://www.jian27.com/html/2659.html


已有 41 条评论

  1. 忐忑与长颈鹿 忐忑与长颈鹿

    up,v7.0明明已经执行成功并且生成完成了但是在软件目录swap文件夹里面找不到[笑哭]

  2. 哈密瓜,数据线顺心 哈密瓜,数据线顺心

    视频是根据时长,还是大小来影响生成速度?

  3. 花瓣阳光 花瓣阳光

    点击exe档后 跳出黑色cmd接口 两秒后消失 什么都没发生 请问yyds知道啥原因吗

    1. 是不是没装运行库?

  4. 花瓣无辜 花瓣无辜

    gtx750ti显卡,cuda11.8和对应版本的cudnn、python、pytorch都装了,选择显卡输出模式还是报错:onnxruntime.capi.onnxruntime_pybind11_state.Fail: [ONNXRuntimeError] : 1 : FAIL : Non-zero status code returned while running Resize node. Name:'Resize_124' Status Message: CUDA error cudaErrorNoKernelImageForDevice:no kernel image is available for execution on the device

    1. 太老的显卡了,也许不支持这么新的驱动,可以试试用cpu跑下看看

  5. 贪玩方冬瓜 贪玩方冬瓜

    大佬帮我看一下是什么原因无法生成视频,感谢
    源图片:F:/lmagematerials/17.jpg,目标视频/图片:C:/Users/Asus/Desktop/000.mp4
    namespace(source_path='F:/lmagematerials/17.jpg', target_path='C:/Users/Asus/Desktop/000.mp4', output_path='./Swap_Out\\17.jpg17088274351.mp4', frame_processors=['face_swapper'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=8, reference_face_position=0, reference_frame_number=0, temp_frame_format='png', temp_frame_quality=80, output_video_encoder='libx264', output_video_quality=100, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='80', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
    namespace(source_path='F:/lmagematerials/17.jpg', target_path='C:/Users/Asus/Desktop/000.mp4', output_path='./Swap_Out\\17.jpg17088274351.mp4', frame_processors=['face_swapper'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=8, reference_face_position=0, reference_frame_number=0, temp_frame_format='png', temp_frame_quality=80, output_video_encoder='libx264', output_video_quality=100, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='80', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
    [FACEFUSION.CORE] Creating temporary resources
    [FACEFUSION.CORE] Extracting frames with 39.00174785610763 FPS
    temp_frames_patterncache_file\facefusion\000\%04d.png
    [FACEFUSION.CORE] Temporary frames not found

    1. 换个模型,升级到最新4.8

  6. 要减肥扯鱼 要减肥扯鱼

    大佬,这个是什么原因呢,视频一直不成功
    源图片:F:/lmagematerials/17.jpg,目标视频/图片:C:/Users/Asus/Desktop/000.mp4
    namespace(source_path='F:/lmagematerials/17.jpg', target_path='C:/Users/Asus/Desktop/000.mp4', output_path='./Swap_Out\\17.jpg17088274351.mp4', frame_processors=['face_swapper'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=8, reference_face_position=0, reference_frame_number=0, temp_frame_format='png', temp_frame_quality=80, output_video_encoder='libx264', output_video_quality=100, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='80', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
    namespace(source_path='F:/lmagematerials/17.jpg', target_path='C:/Users/Asus/Desktop/000.mp4', output_path='./Swap_Out\\17.jpg17088274351.mp4', frame_processors=['face_swapper'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=8, reference_face_position=0, reference_frame_number=0, temp_frame_format='png', temp_frame_quality=80, output_video_encoder='libx264', output_video_quality=100, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='80', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
    [FACEFUSION.CORE] Creating temporary resources
    [FACEFUSION.CORE] Extracting frames with 39.00174785610763 FPS
    temp_frames_patterncache_file\facefusion\000\%04d.png
    [FACEFUSION.CORE] Temporary frames not found

  7. 要减肥扯鱼 要减肥扯鱼

    源图片:F:/lmagematerials/17.jpg,目标视频/图片:C:/Users/Asus/Desktop/000.mp4
    namespace(source_path='F:/lmagematerials/17.jpg', target_path='C:/Users/Asus/Desktop/000.mp4', output_path='./Swap_Out\\17.jpg17088274351.mp4', frame_processors=['face_swapper'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=8, reference_face_position=0, reference_frame_number=0, temp_frame_format='png', temp_frame_quality=80, output_video_encoder='libx264', output_video_quality=100, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='80', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
    namespace(source_path='F:/lmagematerials/17.jpg', target_path='C:/Users/Asus/Desktop/000.mp4', output_path='./Swap_Out\\17.jpg17088274351.mp4', frame_processors=['face_swapper'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=8, reference_face_position=0, reference_frame_number=0, temp_frame_format='png', temp_frame_quality=80, output_video_encoder='libx264', output_video_quality=100, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='80', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
    [FACEFUSION.CORE] Creating temporary resources
    [FACEFUSION.CORE] Extracting frames with 39.00174785610763 FPS
    temp_frames_patterncache_file\facefusion\000\%04d.png
    [FACEFUSION.CORE] Temporary frames not found

    1. 要减肥扯鱼 要减肥扯鱼

      源图片:F:/lmagematerials/17.jpg,目标视频/图片:C:/Users/Asus/Desktop/000.mp4
      namespace(source_path='F:/lmagematerials/17.jpg', target_path='C:/Users/Asus/Desktop/000.mp4', output_path='./Swap_Out\\17.jpg17088274351.mp4', frame_processors=['face_swapper'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=8, reference_face_position=0, reference_frame_number=0, temp_frame_format='png', temp_frame_quality=80, output_video_encoder='libx264', output_video_quality=100, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='80', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
      namespace(source_path='F:/lmagematerials/17.jpg', target_path='C:/Users/Asus/Desktop/000.mp4', output_path='./Swap_Out\\17.jpg17088274351.mp4', frame_processors=['face_swapper'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=8, reference_face_position=0, reference_frame_number=0, temp_frame_format='png', temp_frame_quality=80, output_video_encoder='libx264', output_video_quality=100, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='80', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
      [FACEFUSION.CORE] Creating temporary resources
      [FACEFUSION.CORE] Extracting frames with 39.00174785610763 FPS
      temp_frames_patterncache_file\facefusion\000\%04d.png
      [FACEFUSION.CORE] Temporary frames not found
      这个是什么原因呢

      1. 板栗寒冷 板栗寒冷

        更新到4.5之后 ,图片没事,转视频会提示这个,IFACEFUSION.CORE] Temporary frames not found,未找到临时框架啥的。。。

        1. 缺少模型,可以下载完整包

          1. 贪玩爱高山 贪玩爱高山

            用的就是完整包,换了好几个模型了,都这样

  8. 铃铛孝顺 铃铛孝顺

    大佬我的显卡是N卡是ROG STRIX-GTX1060-A6G-GAMING但是我选择显卡时刚开始正常,过一会儿就报错了报错信息如下
    Error in sys.excepthook:
    Traceback (most recent call last):
    File "roop_face_main_V48_20240217.py", line 1748, in handleException
    AttributeError: 'traceback' object has no attribute 'print_tb'

    Original exception was:
    Traceback (most recent call last):
    File "roop_face_main_V48_20240217.py", line 633, in run
    File "facefusion\core.py", line 191, in cli
    run(program, args_simple, is_live=is_live, signal_in=signal_in, camera_list=camera_list)
    File "facefusion\core.py", line 391, in run
    conditional_process()
    File "facefusion\core.py", line 435, in conditional_process
    process_video(start_time)
    File "facefusion\core.py", line 496, in process_video
    frame_processor_module.process_video(facefusion.globals.source_paths, temp_frame_paths)
    File "F:\tool\AI_huanlian\facefusion\processors\frame\modules\frame_enhancer.py", line 208, in process_video
    frame_processors.multi_process_frames(None, temp_frame_paths, process_frames)
    File "facefusion\processors\frame\core.py", line 87, in multi_process_frames
    future_done.result()
    File "concurrent\futures\_base.py", line 451, in result
    File "concurrent\futures\_base.py", line 403, in __get_result
    File "concurrent\futures\thread.py", line 58, in run
    File "F:\tool\AI_huanlian\facefusion\processors\frame\modules\frame_enhancer.py", line 190, in process_frames
    result_frame = process_frame(
    File "F:\tool\AI_huanlian\facefusion\processors\frame\modules\frame_enhancer.py", line 183, in process_frame
    return enhance_frame(target_vision_frame)
    File "F:\tool\AI_huanlian\facefusion\processors\frame\modules\frame_enhancer.py", line 165, in enhance_frame
    paste_vision_frame, _ = get_frame_processor().enhance(temp_vision_frame)
    File "torch\utils\_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
    File "F:\tool\AI_huanlian\realesrgan\utils.py", line 225, in enhance
    output_img = output_img.data.squeeze().float().cpu().clamp_(0, 1).numpy()
    RuntimeError: CUDA error: misaligned address
    CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
    For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
    Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

    1. 试试更新下显卡驱动

      1. 铃铛孝顺 铃铛孝顺

        好的感谢大佬,我今天升级了显卡驱动还是会报这个错,但是我还了一张照片和视频就能正常运行

  9. 鳗鱼踢背包 鳗鱼踢背包

    can't open file for writing: permission denied

    1. 不要有中文路径或者乱七八糟的符号空格之类

  10. 项链典雅 项链典雅

    点击页面上的问号没有弹出具体的参数说明,有哪里能看到具体的参数说明吗

    1. 鼠标放在问号上不要动,等待2秒会显示具体说明

  11. 暴躁和钢笔 暴躁和钢笔

    *** 捕获到异常 [ONNXRuntimeError] : 3 : NO_SUCHFILE : Load model from D:\BaiduNetdisk\FaceSwapFloder\FaceSwap\models\retinaface_10g.onnx failed:Load model D:\BaiduNetdisk\FaceSwapFloder\FaceSwap\models\retinaface_10g.onnx failed. File doesn't exist
    Error in sys.excepthook:
    Traceback (most recent call last):
    File "roop_face_main_V40_20231124.py", line 1264, in handleException
    AttributeError: 'traceback' object has no attribute 'print_tb'

    Original exception was:
    Traceback (most recent call last):
    File "roop_face_main_V40_20231124.py", line 375, in run
    File "facefusion\core.py", line 90, in cli
    run(program, args_simple,is_live=is_live,signal_in=signal_in,camera_list=camera_list)
    File "facefusion\core.py", line 232, in run
    conditional_process()
    File "facefusion\core.py", line 267, in conditional_process
    conditional_set_face_reference()
    File "facefusion\core.py", line 283, in conditional_set_face_reference
    reference_face = get_one_face(reference_frame, facefusion.globals.reference_face_position)
    File "facefusion\face_analyser.py", line 228, in get_one_face
    many_faces = get_many_faces(frame)
    File "facefusion\face_analyser.py", line 243, in get_many_faces
    faces = extract_faces(frame)
    File "facefusion\face_analyser.py", line 106, in extract_faces
    bbox_list, kps_list, score_list = detect_with_retinaface(temp_frame, temp_frame_height, temp_frame_width, face_detector_height, face_detector_width, ratio_height, ratio_width)
    File "facefusion\face_analyser.py", line 115, in detect_with_retinaface
    face_detector = get_face_analyser().get('face_detector')
    File "facefusion\face_analyser.py", line 58, in get_face_analyser
    face_detector = onnxruntime.InferenceSession(MODELS.get('face_detector_retinaface').get('path'), providers = facefusion.globals.execution_providers)
    File "onnxruntime\capi\onnxruntime_inference_collection.py", line 419, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
    File "onnxruntime\capi\onnxruntime_inference_collection.py", line 452, in _create_inference_session
    sess = C.InferenceSession(session_options, self._model_path, True, self._read_config_from_model)
    onnxruntime.capi.onnxruntime_pybind11_state.NoSuchFile: [ONNXRuntimeError] : 3 : NO_SUCHFILE : Load model from D:\BaiduNetdisk\FaceSwapFloder\FaceSwap\models\retinaface_10g.onnx failed:Load model

  12. 山水纯真 山水纯真

    源图片:H:/ai/1.png,目标视频/图片:H:/ai/IMG_7407(20231123-181057).JPG
    namespace(source_path='H:/ai/1.png', target_path='H:/ai/IMG_7407(20231123-181057).JPG', output_path='./Swap_Out\\1.png17007383971_HD.png', frame_processors=['face_swapper', 'face_enhancer'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=4, reference_face_position=0, reference_frame_number=0, temp_frame_format='jpg', temp_frame_quality=100, output_video_encoder='libx264', output_video_quality=90, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='100', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
    namespace(source_path='H:/ai/1.png', target_path='H:/ai/IMG_7407(20231123-181057).JPG', output_path='./Swap_Out\\1.png17007383971_HD.png', frame_processors=['face_swapper', 'face_enhancer'], keep_fps=True, skip_audio=False, keep_temp=False, max_memory=16, execution_providers=['cuda'], execution_thread_count=4, reference_face_position=0, reference_frame_number=0, temp_frame_format='jpg', temp_frame_quality=100, output_video_encoder='libx264', output_video_quality=90, ui_layouts=['default'], face_recognition='reference', face_analyser_direction='left-right', face_analyser_age=None, face_analyser_gender=None, reference_face_distance=1500, trim_frame_start=None, trim_frame_end=None, execution_queue_count=1, zoom_scale=1, zoom_tile=512, zoom_fp16=True, skip_download=True, headless=True, face_enhancer_model='gfpgan_1.4', face_enhancer_blend='100', face_swapper_model='inswapper_128', frame_enhancer_model='realesrgan_x2plus', frame_enhancer_blend='100', output_image_quality=80)
    [FACEFUSION.FRAME_PROCESSOR.FACE_SWAPPER] Processing处理
    2023-11-23 19:20:47.6006543 [E:onnxruntime:, sequential_executor.cc:514 onnxruntime::ExecuteKernel] Non-zero status code returned while running Resize node. Name:'Resize_124' Status Message: CUDA error cudaErrorNoKernelImageForDevice:no kernel image is available for execution on the device
    *** 捕获到异常 [ONNXRuntimeError] : 1 : FAIL : Non-zero status code returned while running Resize node. Name:'Resize_124' Status Message: CUDA error cudaErrorNoKernelImageForDevice:no kernel image is available for execution on the device
    Error in sys.excepthook:
    Traceback (most recent call last):
    File "roop_face_main_V30_20231014.py", line 1009, in handleException
    AttributeError: 'traceback' object has no attribute 'print_tb'

    Original exception was:
    Traceback (most recent call last):
    File "roop_face_main_V30_20231014.py", line 325, in run
    File "facefusion\core.py", line 115, in cli
    run(program, args_simple,is_live=is_live,signal_in=signal_in,camera_list=camera_list)
    File "facefusion\core.py", line 233, in run
    conditional_process()
    File "facefusion\core.py", line 281, in conditional_process
    process_image()
    File "facefusion\core.py", line 293, in process_image
    frame_processor_module.process_image(facefusion.globals.source_path, facefusion.globals.output_path,
    File "H:\aaai\facefusion\processors\frame\modules\face_swapper.py", line 147, in process_image
    source_face = get_one_face(read_static_image(source_path))
    File "facefusion\face_analyser.py", line 31, in get_one_face
    many_faces = get_many_faces(frame)
    File "facefusion\face_analyser.py", line 46, in get_many_faces
    faces = get_face_analyser().get(frame)
    File "insightface\app\face_analysis.py", line 59, in get
    File "insightface\model_zoo\retinaface.py", line 224, in detect
    File "insightface\model_zoo\retinaface.py", line 152, in forward
    File "onnxruntime\capi\onnxruntime_inference_collection.py", line 220, in run
    return self._sess.run(output_names, input_feed, run_options)
    onnxruntime.capi.onnxruntime_pybind11_state.Fail: [ONNXRuntimeError] : 1 : FAIL : Non-zero status code returned while running Resize node. Name:'Resize_124' Status Message: CUDA error cudaErrorNoKernelImageForDevice:no kernel image is available for execution on the device
    什么情况

  13. 现实和鱼 现实和鱼

    ADM 就是A卡吧

  14. nigerdxg nigerdxg

    感谢分享

  15. 冰淇淋坚强 冰淇淋坚强

    为何楼主你这么优秀,为什么 ,为什么

  16. 勤恳爱人生 勤恳爱人生

    你好,视频人物动的时候会跑出一堆不一样的脸,有男有女,停下来也还是原来的样子,这个怎么办

    1. 最好是只有单人,过于复杂的是不行的。。。

      1. 火星上笑灰狼 火星上笑灰狼

        我照片和视频都是单人,视频出来还是原视频的人脸,视频只是跳舞,没有很复杂的动作

        1. 可以试试换别的视频看看行不行

    2. 不要动头部运动幅度过大

  17. 缘分无奈 缘分无奈

    如有疑问,请留言告知。一支穿云箭,千军万马来相见;两副忠义胆,刀山火海提命现。

    1. 标致迎大白 标致迎大白

      好吧谢谢你,还有一个,我已经装了运行库,但用显卡跑会出现 File "roop_face_main_V17_20230730.py", line 152, in run
      File "roop\core.py", line 305, in run
      File "roop\core.py", line 258, in start
      File "D:\KuaiwanGames\Games\AIV1.8\roop\processors\frame\face_swapper.py", line 100, in process_video
      roop.processors.frame.core.process_video(source_path, temp_frame_paths, process_frames)
      File "roop\processors\frame\core.py", line 73, in process_video
      multi_process_frame(source_path, frame_paths, process_frames, lambda: update_progress(progress))
      File "roop\processors\frame\core.py", line 51, in multi_process_frame
      future.result()
      File "concurrent\futures\_base.py", line 451, in result
      File "concurrent\futures\_base.py", line 403, in __get_result
      File "concurrent\futures\thread.py", line 58, in run
      File "D:\KuaiwanGames\Games\AIV1.8\roop\processors\frame\face_swapper.py", line 77, in process_frames
      source_face = get_one_face(cv2.imread(source_path))
      File "roop\face_analyser.py", line 30, in get_one_face
      many_faces = get_many_faces(frame)
      File "roop\face_analyser.py", line 41, in get_many_faces
      return get_face_analyser().get(frame)
      File "insightface\app\face_analysis.py", line 75, in get
      model.get(img, face)
      File "insightface\model_zoo\arcface_onnx.py", line 67, in get
      face.embedding = self.get_feat(aimg).flatten()
      File "insightface\model_zoo\arcface_onnx.py", line 84, in get_feat
      net_out = self.session.run(self.output_names, {self.input_name: blob})[0]
      File "onnxruntime\capi\onnxruntime_inference_collection.py", line 217, in run
      return self._sess.run(output_names, input_feed, run_options)
      onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXR

      1. 是什么显卡?A卡的话 不行 要选CPU

    2. 火星上笑灰狼 火星上笑灰狼

      另一个视频也不行,而且导出来从20秒变成了18分钟,呜呜

      1. 标致迎大白 标致迎大白

        电脑上标签贴的是N卡

        1. 可以用鲁大师具体看看是什么显卡。如果显卡不行就选CPU

      2. 我没遇到过这个情况

添加新评论