{"id":1615,"date":"2020-02-28T23:54:38","date_gmt":"2020-02-28T15:54:38","guid":{"rendered":"https:\/\/www.deepfaker.xyz\/?p=1615"},"modified":"2020-05-27T21:04:01","modified_gmt":"2020-05-27T13:04:01","slug":"deepfacelab-%e8%bf%91%e6%9c%9f%e6%9b%b4%e6%96%b0%e6%a6%82%e8%a6%81%ef%bc%8c%e6%95%b4%e8%84%b8%e6%9d%a5%e8%a2%ad%ef%bc%81","status":"publish","type":"post","link":"https:\/\/www.deepfaker.xyz\/?p=1615","title":{"rendered":"DeepFaceLab \u8fd1\u671f\u66f4\u65b0\u6982\u8981\uff0c\u201c\u6574\u8138\u201d\u6765\u88ad\uff01"},"content":{"rendered":"

2020\u5e742\u67083\u53f7\u4e4b\u540e\uff0cIperov\u540c\u5fd7\u6c89\u6d78\u4e86\u4e00\u5c0f\u6bb5\u65f6\u95f4\u3002\u4e0d\u8fc7\uff0c\u6ca1\u591a\u4e45\u53c8\u8fdb\u5165\u4e86\u75af\u72c2\u66f4\u65b0\u7684\u72b6\u6001\u3002\u76ee\u524d\u5df2\u7ecf\u53d1\u5e03\u4e8628\u53f7\uff08\u4eca\u5929\uff09\u7684\u7248\u672c\u3002<\/p>\n

\"\"
\n\u6211\u8fd9\u91cc\u5148\u62ce\u51e0\u4e2a\u91cd\u70b9\u3002<\/p>\n

1. \u4fee\u590d\u773c\u795e\u8be1\u5f02\u6216\u8005\u4e71\u98d8\u7684\u95ee\u9898\u3002<\/p>\n

\u4f7f\u7528DFL2\u540e\uff0c\u4e0d\u5c11\u4eba\u53d1\u73b0\u773c\u795e\u4e0a\u7684\u95ee\u9898\uff0c\u5177\u4f53\u7684\u4f53\u73b0\u6709\u773c\u795e\u5b8c\u5168\u4e0d\u540c\u6b65\uff08\u5f53\u7136\u6709\u4e9b\u4eba\u662f\u7d20\u6750\u4e0d\u591f\uff09\uff0c\u8fd8\u6709\u4e00\u79cd\u662f\u773c\u795e\u5f88\u8be1\u5f02\uff0c\u611f\u89c9\u53d8\u5f02\u4e86\u4e00\u6837\u3002<\/p>\n

\u9488\u5bf9\u8fd9\u4e24\u79cd\u60c5\u51b5\uff0c\u8f6f\u4ef6\u8fdb\u884c\u4e86\u6539\u8fdb\u3002\u5177\u4f53\u6709\u591a\u5927\u6539\u5584\uff0c\u8fd8\u6709\u5f85\u89c2\u5bdf\u3002<\/p>\n

2. AI\u6362\u8138+AE\u7279\u6548\u7684\u201c\u6574\u8138\u201d\u6a21\u578b<\/p>\n

\"\"<\/p>\n

\u7ee7\u534a\u8138\uff0c\u5168\u8138\uff0c\u4e2d\u8138\uff0c\u4e4b\u540e\u53c8\u6dfb\u52a0\u4e86\u65b0\u7684\u4e00\u79cd\u8138\u578bWhole_Face\uff0c\u8fd9\u4e2a\u5176\u5b9e\u53ef\u4ee5\u7b80\u79f0\u4e3a\u201c\u5168\u8138\u201d \uff0c\u4f46\u662f\u56e0\u4e3a\u4e4b\u524d\u5df2\u7ecf\u6709\u4e00\u79cdFUll_FACE , \u4e3a\u4e86\u533a\u522b\u6211\u89c9\u5f97\u7ffb\u8bd1\u6210\u201c\u6574\u8138\u201d \u6bd4\u8f83\u597d\u3002\u8fd9\u4e2a\u7c7b\u578b\uff0c\u6700\u5927\u7684\u529f\u80fd\u5c31\u662f\u89e3\u51b3\u989d\u5934\u7684\u95ee\u9898\u3002<\/p>\n

\u4f46\u662f…\u8fd9\u662f\u4e00\u4e2a\u534a\u6210\u54c1\uff0c\u9700\u8981\u7ed3\u5408AE\u624d\u80fd\u505a\u51fa\u5b8c\u6574\u7684\u6548\u679c\u3002\u7b80\u7565\u6559\u7a0b\u53c2\u89c1\u4e0a\u56fe\uff0c\u5177\u4f53\u4f7f\u7528\u65b9\u6cd5\uff0c\u53ef\u4ee5\u5173\u6ce8\u540e\u7eed\u6587\u7ae0\u3002<\/p>\n

3. 256\u63d0\u5347\u5230512\u3002<\/p>\n

\"\"<\/p>\n

\u56e0\u4e3a\u51fa\u4e86\u5168\u8138\u4e4b\u540e\uff0c\u5f15\u53d1\u7684\u8774\u8776\u6548\u5e94\u3002\u539f\u5148\u6211\u4eec\u63d0\u53d6\u7684\u56fe\u7247\u90fd\u662f256×256\u50cf\u7d20\uff0c\u73b0\u5728\u56e0\u4e3a\u9700\u8981\u5904\u7406\u6574\u4e2a\u8138\u90e8\uff0c\u6240\u4ee5\u63d0\u5347\u4e86\u5934\u50cf\u5c3a\u5bf8\uff0c\u6269\u5230\u4e86512×512\u3002\u540c\u65f6\u628a\u8bad\u7ec3\u6a21\u578b\u50cf\u7d20\u7684\u6700\u5927\u503c\u4e5f\u6539\u5230\u4e86512\uff0832G,V100\u542c\u5b8c\u540e\u745f\u745f\u53d1\u6296\uff0c\u8bf4\u4e86\u53e5\uff1a\u544a\u8f9e\uff09\u3002 \u867d\u7136\u591a\u4e2a\u5730\u65b9\u7684size\u90fd\u6539\u5927\u4e86\uff0c\u4f46\u662f\u5e76\u4e0d\u610f\u5473\u7740\u4f60\u7684\u663e\u5361\u80fd\u8dd1\u51fa\u66f4\u597d\u7684\u6548\u679c…<\/p>\n

\u8fd8\u6709\u5b89\u88c5\u5305\u662f\u8d8a\u6765\u8d8a\u5927\u4e86…<\/p>\n

\u6700\u8fd1\u4e09\u6b21\u66f4\u65b0\u7684\u4e3b\u8981\u5185\u5bb9\u5e94\u8be5\u5c31\u662f\u8fd9\u4e9b\u4e86\uff0c\u5177\u4f53\u66f4\u65b0\u8bb0\u5f55\u5982\u4e0b\uff1a<\/p>\n

2020\u5e742\u670828\u65e5
\nExtractor:
\n\u63d0\u53d6<\/p>\n

image size for all faces is now 512
\n\u63d0\u53d6\u7684\u56fe\u7247\u5c3a\u5bf8\u4ece256\u63d0\u5347\u5230512<\/p>\n

fix RuntimeWarning during the extraction process
\n\u4fee\u590d\u63d0\u53d6\u8fc7\u7a0b\u4e2d\u51fa\u73b0RuntimeWarning\u7684\u95ee\u9898\u3002<\/p>\n

SAEHD:
\nmax resolution is now 512
\n\u6a21\u578b\u53c2\u6570\u4e2d\uff0c\u6700\u9ad8\u50cf\u7d20\u6539\u6210512\uff0832G V100\u540c\u6837\u8868\u793a\u73a9\u4e0d\u8d77\uff09<\/p>\n

fix hd arhitectures. Some decoder\u2019s weights haven\u2019t trained before.
\n\u4fee\u590dHD\u67b6\u6784\uff0c\u4e4b\u524d\u4e00\u4e9b\u89e3\u7801\u5668\u6743\u91cd\u6ca1\u6709\u88ab\u8bad\u7ec3\u5230\u3002<\/p>\n

new optimized training:
\n\u65b0\u7684\u4f18\u5316\u8bad\u7ec3<\/p>\n

for every <batch_size*16> samples,
\n\u5bf9\u4e8e\u6bcf\u4e2a<batch_size*16>\u6837\u672c\uff0c<\/p>\n

model collects <batch_size> samples with the highest error and learns them again
\n\u6a21\u578b\u6536\u96c6\u8bef\u5dee\u6bd4\u8f83\u5927\u7684\u6837\u4f8b\u8fdb\u884c\u91cd\u65b0\u5b66\u4e60\u3002<\/p>\n

therefore hard samples will be trained more often
\n\u56e0\u6b64\uff0c\u6bd4\u8f83\u96be\u7684\u6837\u672c\u88ab\u8bad\u7684\u66f4\u72e0\u3002<\/p>\n

\u2018models_opt_on_gpu\u2019 option is now available for multigpus (before only for 1 gpu)
\n\u201cmodels_opt_on_gpu\u201d\u9009\u9879\u73b0\u5728\u53ef\u7528\u4e8e\u591a\u663e\u5361\uff08\u4ee5\u524d\u4ec5\u7528\u5355\u663e\u5361\uff09<\/p>\n

fix \u2018autobackup_hour\u2019
\n\u4fee\u590d\u201c\u81ea\u52a8\u5907\u4efd\u65f6\u95f4\u201d<\/p>\n

2020\u5e742\u670823\u65e5
\nSAEHD: pretrain option is now available for whole_face type
\nSAEHD: \u9884\u8bad\u7ec3\u652f\u6301\u201c\u6574\u8138\u201d<\/p>\n

fix sort by abs difference
\n\u4fee\u590dabs\u5dee\u5f02\u6392\u5e8f<\/p>\n

fix sort by yaw\/pitch\/best for whole_face\u2019s
\n\u4fee\u590d\u6574\u8138\u7684yaw\/pitch\/best\u6392\u5e8f\u3002<\/p>\n

2020\u5e742\u670821\u65e5<\/p>\n

Trainer: decreased time of initialization
\n\u8bad\u7ec3\uff1a\u51cf\u5c11\u521d\u59cb\u5316\u65f6\u95f4<\/p>\n

Merger: fixed some color flickering in overlay+rct mode
\n\u5408\u6210\uff1a\u4fee\u590doverlay+rct\u6a21\u5f0f\u4e0b\u7684\u4e00\u4e9b\u95ea\u7f29\u7684\u95ee\u9898\u3002<\/p>\n

SAEHD:
\nadded option Eyes priority (y\/n)
\n\u6dfb\u52a0\u773c\u90e8\u4f18\u5148\u9009\u9879<\/p>\n

Helps to fix eye problems during training like \u201calien eyes\u201d
\n\u6709\u52a9\u4e8e\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u89e3\u51b3\u201c\u5f02\u5f62\u773c\u201d<\/p>\n

and wrong eyes direction ( especially on HD architectures )
\n\u548c\u773c\u795e\u4e0d\u5bf9\u7684\u95ee\u9898\uff08\u5c24\u5176\u662f\u5728HD\u67b6\u6784\u4e0a\uff09<\/p>\n

by forcing the neural network to train eyes with higher priority.
\n\u901a\u8fc7\u5f3a\u5236\u795e\u7ecf\u7f51\u7edc\u4ee5\u66f4\u9ad8\u7684\u4f18\u5148\u7ea7\u6765\u8bad\u7ec3\u773c\u775b\u3002<\/p>\n

before\/after https:\/\/i.imgur.com\/YQHOuSR.jpg
\n\u4e4b\u524d\/\u4e4b\u540e<\/p>\n

added experimental face type \u2018whole_face\u2019
\n\u6dfb\u52a0\u8bd5\u9a8c\u6027\u8138\u578b\u201c\u6574\u8138\u201d\u3002<\/p>\n

Basic usage instruction: https:\/\/i.imgur.com\/w7LkId2.jpg
\n\u57fa\u672c\u7528\u6cd5\u8bf4\u660e\uff1a<\/p>\n

\u2018whole_face\u2019 requires skill in Adobe After Effects.
\n\u6574\u8138\u9700\u8981 AE\u6280\u80fd<\/p>\n

For using whole_face you have to extract whole_face\u2019s by using
\n\u4f7f\u7528\u6574\u8138\u4f60\u5fc5\u987b\u63d0\u53d6\u6574\u8138\u56fe\u7247\uff0c\u901a\u8fc7\u4ee5\u4e0b\u4e24\u4e2a\u811a\u672c\u3002<\/p>\n

4) data_src extract whole_face
\nand
\n5) data_dst extract whole_face
\nImages will be extracted in 512 resolution, so they can be used for regular full_face\u2019s and half_face\u2019s.
\n\u56fe\u7247\u4f1a\u6309512\u50cf\u7d20\u8fdb\u884c\u63d0\u53d6\u3002\u4ed6\u4eec\u540c\u6837\u4e5f\u9002\u7528\u4e8e\u5e38\u89c4\u7684\u5168\u8138\u548c\u534a\u8138\u3002<\/p>\n

\u2018whole_face\u2019 covers whole area of face include forehead in training square,
\n\u6574\u8138\u6db5\u76d6\u8138\u90e8\u7684\u6574\u4e2a\u533a\u57df\u5305\u542b\u989d\u5934\u3002<\/p>\n

but training mask is still \u2018full_face\u2019
\n\u4f46\u662f\u8bad\u7ec3\u906e\u7f69\u4f9d\u65e7\u662f\u5168\u8138<\/p>\n

therefore it requires manual final masking and composing in Adobe After Effects.
\n\u56e0\u6b64\u9700\u8981\u4f60\u901a\u8fc7AE\u624b\u52a8\u906e\u7f69\u548c\u5408\u6210\u3002<\/p>\n

added option \u2018masked_training\u2019
\n\u6dfb\u52a0\u906e\u7f69\u8bad\u7ec3<\/p>\n

This option is available only for \u2018whole_face\u2019 type.
\n\u8fd9\u4e2a\u9009\u9879\u4ec5\u9002\u7528\u4e8e\u6574\u8138<\/p>\n

Default is ON.
\n\u9ed8\u8ba4\u5f00\u542f<\/p>\n

Masked training clips training area to full_face mask,
\n\u906e\u7f69\u8bad\u7ec3\u53ef\u4ee5\u628a\u8bad\u7ec3\u533a\u57df\u526a\u8f91\u5230\u5168\u8138\u906e\u7f69\u3002<\/p>\n

thus network will train the faces properly.
\n\u8fd9\u6837\u7f51\u7edc\u5c31\u53ef\u4ee5\u6b63\u786e\u8bad\u7ec3\u8138\u90e8<\/p>\n

When the face is trained enough, disable this option to train all area of the frame.
\n\u5f53\u8bad\u7ec3\u8db3\u591f\u591a\u7684\u65f6\u5019\uff0c\u5173\u95ed\u8fd9\u4e2a\u9009\u9879\uff0c\u8ba9\u7f51\u7edc\u53bb\u8bad\u7ec3\u6240\u6709\u533a\u57df\u3002<\/p>\n

Merge with \u2018raw-rgb\u2019 mode, then use Adobe After Effects to manually mask, tune color, and compose whole face include forehead.
\n\u4f7f\u7528Raw-rgb\u6a21\u5f0f\u5408\u6210\uff0c\u7136\u540e\u4f7f\u7528AE\u624b\u52a8\u906e\u7f69\u8c03\u6574\u989c\u8272\uff0c\u7136\u540e\u5408\u6210\uff0c\u8fd9\u6837\u989d\u5934\u4e5f\u80fd\u88ab\u6362\u6389\u3002<\/p>\n

\u66f4\u591a\u66f4\u65b0\u5185\u5bb9\u8bf7\u770b\uff1ahttps:\/\/www.deepfaker.xyz\/?page_id=1604
\n\u6700\u65b0\u7248\u89c1659480116\u7fa4\u6587\u4ef6\uff0c\u636e\u8bf4\u8f6c\u53d1\u4e00\u4e0b\u53ef\u4ee5\u6781\u901f\u4e0b\u8f7d\u54e6\uff01<\/p>\n","protected":false},"excerpt":{"rendered":"

2020\u5e742\u67083\u53f7\u4e4b\u540e\uff0cIperov\u540c\u5fd7\u6c89\u6d78\u4e86\u4e00\u5c0f\u6bb5\u65f6\u95f4\u3002\u4e0d\u8fc7\uff0c\u6ca1\u591a\u4e45\u53c8\u8fdb\u5165\u4e86\u75af\u72c2\u66f4\u65b0\u7684\u72b6\u6001\u3002\u76ee\u524d\u5df2\u7ecf\u53d1\u5e03\u4e862 […]<\/p>\n","protected":false},"author":1,"featured_media":1617,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[4,3],"tags":[14,12,10],"_links":{"self":[{"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=\/wp\/v2\/posts\/1615"}],"collection":[{"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1615"}],"version-history":[{"count":3,"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=\/wp\/v2\/posts\/1615\/revisions"}],"predecessor-version":[{"id":1882,"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=\/wp\/v2\/posts\/1615\/revisions\/1882"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=\/wp\/v2\/media\/1617"}],"wp:attachment":[{"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1615"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1615"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.deepfaker.xyz\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1615"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}