【人脸图清晰化神器】GFPGAN_GPEN_CodeFormer 3合1人像增强工具集

slib

报错RuntimeError: CUDA out of memory.    重置过缓存之类的了，而且明明显存根本没用完

slib

然后这个提升完画质，直接就可以放到aligned里用了？还是依旧要运行一遍face extract

leeso

slib 发表于 2021-10-11 19:26
然后这个提升完画质，直接就可以放到aligned里用了？还是依旧要运行一遍face extract ...

这个报错说明你显存不够啊
更新GUI界面了，你研究下
支持优化完毕直接放到aligned使用，前提是原图是切过脸的

ugtqqaikv854801

感谢分享

slib

leeso 发表于 2021-10-11 21:29
这个报错说明你显存不够啊
更新GUI界面了，你研究下
支持优化完毕直接放到aligned使用，前提是原图是切过 ...

他这个很奇怪，我在输入文件夹里放了100张，给我输出60张，我在输入文件夹里只留下还没输出的那些照片，再运行一次程序一张都不输出，一样的报错

DFL8938o9

感谢大神 有图形界面更方便了

772453465

好东西值得收藏

slib

D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\functional.py:718: UserWarning: Named tensors and all their associated APIs are an experimental feature and subject to change. Please do not use them for anything important until they are released as stable. (Triggered internally at  ..\c10/core/TensorImpl.h:1156.)
  return torch.max_pool2d(input, kernel_size, stride, padding, dilation, ceil_mode)
Traceback (most recent call last):
  File "face_enhancement_BFR512.py", line 101, in <module>
    img, orig_faces, enhanced_faces = faceenhancer.process(im)
  File "face_enhancement_BFR512.py", line 43, in process
    facebs, landms = self.facedetector.detect(img)
  File "D:\GPEN-GFPGAN-MSOLD\GPEN\retinaface\retinaface_detection.py", line 67, in detect
    loc, conf, landms = self.net(img)  # forward pass
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\modules\module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "D:\GPEN-GFPGAN-MSOLD\GPEN\retinaface\facemodels\retinaface.py", line 108, in forward
    out = self.body(inputs)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\modules\module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torchvision\models\_utils.py", line 62, in forward
    x = module(x)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\modules\module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\modules\container.py", line 139, in forward
    input = module(input)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\modules\module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torchvision\models\resnet.py", line 136, in forward
    identity = self.downsample(x)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\modules\module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\modules\container.py", line 139, in forward
    input = module(input)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\modules\module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\modules\batchnorm.py", line 178, in forward
    self.eps,
  File "D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\functional.py", line 2282, in batch_norm
    input, weight, bias, running_mean, running_var, training, momentum, eps, torch.backends.cudnn.enabled
RuntimeError: CUDA out of memory. Tried to allocate 636.00 MiB (GPU 0; 4.00 GiB total capacity; 1.88 GiB already allocated; 282.65 MiB free; 2.52 GiB reserved in total by PyTorch)
请按任意键继续. . .   我开着任务管理器看gpu使用情况，根本没使用 然后使用新的那个gui界面cpu使用率也没上过50%，内存占用率没变化稳定在50，处理进度不动，清理过缓存了

无敌小金鱼

这个真心比较好的工具，值得花再多石头

leeso

slib 发表于 2021-10-12 11:45
D:\GPEN-GFPGAN-MSOLD\_internal\Python37\lib\site-packages\torch\nn\functional.py:718: UserWarning: N ...

显存太小 ，试试换成256的模型