primepake/wav2lip_288x288

Issues

• how to use syncnet_python and training steps

  #124 opened 10 months ago by 1129571
  13

• train_syncnet_sam.py加载无反应

  #110 opened a year ago by fzuqxt
  9

• AttributeError: module 'face_detection' has no attribute 'FaceAlignment'

  #158 opened 2 months ago by wxc1207
  0

• why train_syncnet_sam.py set img_size=384?

  #157 opened 3 months ago by qingshuiyuyu
  1

• dataset

  #147 opened 7 months ago by see2run
  1

• how to prepare training dataset

  #156 opened 3 months ago by dxLu
  0

• 大家有遇到过执行train_syncnet_sam.py后出现的张量维度不一致问题吗：tensor a (1024) must match the size of tensor b (9216)，如图

  #155 opened 3 months ago by yuanmaitian
  1

• 出售一份自己人工手动下载人脸数字人的数据集原始视频 ，视频分辨率1080-4k口播视频，可用于训练wav2lip-192，wav2lip-288，wav2lip-384，wav2lip-512，DINet-256，NeRF等模型。微信douzijun1999 请注明来意

  #154 opened 3 months ago by dousang1999
  0

• 每次训练停止，都从头开始，怎么修改能继续上次断开的地方开始

  #153 opened 3 months ago by zn123
  0

• also around the 0.69 on 240000 steps

  #118 opened a year ago by cncbec
  3

• The amplitude of the mouth shape becomes smaller with custom dataset

  #152 opened 3 months ago by zhangziwenHIT
  0

• Model copying reference frames as it is

  #122 opened 10 months ago by mudassirkhan19
  6

• Why can’t training start?

  #142 opened 8 months ago by piwawa
  4

• Syncnet loss does not converge

  #146 opened 7 months ago by kavita-gsphk
  22

• structure and arrangement of dataset before executing any scrip with flow of script execution!!!!!!

  #151 opened 5 months ago by vpssa
  0

• Use HuBERT features to train SyncNet, the loss does not converge.

  #150 opened 6 months ago by hnsywangxin
  2

• DINet

  #148 opened 7 months ago by duyquang099
  1

• 这个和普通的easyw字幕交换网站lip有什么区别

  #149 opened 7 months ago by tuantra-dh
  0

• DINet implementation

  #136 opened 9 months ago by ozzieakg
  1

• wav2lip sam

  #128 opened 10 months ago
  1

• Training failed. The lip shape of a character cannot change according to changes in speech

  #145 opened 7 months ago by Liming-belief
  6

• Find friends who are training models and share ideas with them.Welcome

  #133 opened 9 months ago by LiuJiaqi0505
  3

• Why my train loss after introducing sync loss?

  #140 opened 8 months ago by Marskly
  4

• How to train

  #141 opened 8 months ago by see2run
  6

• Generated bottom half face always blur.

  #144 opened 7 months ago by wangjio
  2

• do inference

  #143 opened 7 months ago by see2run
  0

• Hi sir, I am a beginner and I would like to inquire whether I should prepare a video of no less than 288 or a video of 384

  #132 opened 8 months ago by SakuraMaiii
  0

• Train syncnet use SyncNet_color_384 but train wav2lip use SyncNet_color?

  #134 opened 9 months ago by thinkyang88
  1

• What indicator represents the end of training hq_wav2lip_sam_train？

  #139 opened 8 months ago by mrhanCH
  4

• train_syncnet_sam.py is not using GPU (RTX 4090)

  #138 opened 8 months ago by meng-x
  1

• LOSS stuck arount 0.69

  #114 opened a year ago by ashishjamarkattel
  7

• When I use hq_wav2lip_sam_train.py。

  #135 opened 9 months ago by yuan7245
  3

• Syncnet training? Chilcken and egg?

  #119 opened a year ago by sylyt62
  4

• High resolution dataset

  #131 opened 10 months ago by akshatgarg99
  1

• video clips length

  #137 opened 9 months ago by hnsywangxin
  0

• the input of lpips loss

  #130 opened 10 months ago by taozhiqi
  1

• Hi Brother

  #129 opened 10 months ago by langzizhixin
  2

• Wav2Lip_SAM freeze_audio_encoder

  #127 opened 10 months ago by taozhiqi
  1

• Should this be `Wav2Lip_SAM` or `Wav2Lip_384`?

  #126 opened 10 months ago by albyho
  0

• 老哥，你为嘛把项目名称改来改去？

  #123 opened 10 months ago by prometheus-alien
  5

• Missing License

  #125 opened 10 months ago by ntakouris
  3

• train_syncnet history，waiting the result

  #108 opened a year ago by zhangziliang04
  3

• During inference, how to remove an additional mask layer on the face?

  #121 opened a year ago by keepWhere
  0

• Is 288 Sam supported? I changed the image size of syncnet to 288 and trained it, but found that the second step does not support 288

  #120 opened a year ago by einsqing
  0

• Help to test the data from open-sourced dataset

  #117 opened a year ago by Nyquist0
  0

• instruction

  #116 opened a year ago by Citramones
  0

• RuntimeError: Error(s) in loading state_dict for SyncNet_color:

  #115 opened a year ago by PhilanthropistBright
  1

• when I runnning the preprocess.py, the process unresponsive

  #112 opened a year ago by fqhwwlzzbh
  0

• inference no working

  #113 opened a year ago by dezei547
  0

• Is there a pre-trained model？

  #111 opened a year ago by 397688551
  0