MRzzm / DINet

The source code of "DINet: deformation inpainting network for realistic face visually dubbing on high resolution video."
956 stars 169 forks source link

用自己的视频数据测试,推理时视频帧数与openface 生成的csv的帧数不一致 #27

Open muximuxi opened 1 year ago

muximuxi commented 1 year ago

您好! 用自己的视频数据测试,推理时视频帧数与openface 生成的csv的帧数不一致,请问怎么解决?是视频源有问题吗?但是换了好几个视频都是这样: The video_landmark_data.shape: 249 aligning frames with driving audio len_video_frames: 250 Traceback (most recent call last): File "inference.py", line 67, in raise ('video frames are misaligned with detected landmarks') TypeError: exceptions must derive from BaseException

SetoDon commented 1 year ago

I also have the same problem, how to solve it.

Laixinsz commented 1 year ago

抛弃多余的帧数,截取openface检测的帧数,可以继续生成,但是不知道会不会影响效果,用dinet自带的示例视频用openface生成csv文件不会出现这种帧数不一致的情况,不知道设么原因,自带的示例视频是29fps/s的

ZYBOBO commented 1 year ago

抛弃多余的帧数,截取openface检测的帧数,可以继续生成,但是不知道会不会影响效果,用dinet自带的示例视频用openface生成csv文件不会出现这种帧数不一致的情况,不知道设么原因,自带的示例视频是29fps/s的

可以把帧率调成25fps/s的试试

ZYBOBO commented 1 year ago

抛弃多余的帧数,截取openface检测的帧数,可以继续生成,但是不知道会不会影响效果,用dinet自带的示例视频用openface生成csv文件不会出现这种帧数不一致的情况,不知道设么原因,自带的示例视频是29fps/s的

请教一下,利用OpenFace得到.csv文件的完整命令是啥?谢谢~

Laixinsz commented 1 year ago

docker exec -it openface /home/openface-build/build/bin/FeatureExtraction -f /tmp/openface/fc7496f57248684429995b45ecfebd0f.mp4 -2Dfp -wild -tracked -nomask -out_dir /tmp/openface/fc7496f57248684429995b45ecfebd0fcsv 我用docker安装的。不过经常会报 0% 10% BLAS : Program is Terminated. Because you tried to allocate too many memory regions. BLAS : Program is Terminated. Because you tried to allocate too many memory regions. BLAS : Program is Terminated. Because you tried to allocate too many memory regions. 但是csv是生成了,可能没有完整生成,生成了部分帧。

抛弃多余的帧数,截取openface检测的帧数,可以继续生成,但是不知道会不会影响效果,用dinet自带的示例视频用openface生成csv文件不会出现这种帧数不一致的情况,不知道设么原因,自带的示例视频是29fps/s的

请教一下,利用OpenFace得到.csv文件的完整命令是啥?谢谢~

laishujie commented 10 months ago

您好! 用自己的视频数据测试,推理时视频帧数与openface 生成的csv的帧数不一致,请问怎么解决?是视频源有问题吗?但是换了好几个视频都是这样: The video_landmark_data.shape: 249 aligning frames with driving audio len_video_frames: 250 Traceback (most recent call last): File "inference.py", line 67, in raise ('video frames are misaligned with detected landmarks') TypeError: exceptions must derive from BaseException

如果你是使用应用程序的话,尝试换一下检测器试一试。 Snipaste_2023-11-17_16-45-16

zxc67373 commented 9 months ago

有可能是因为没有检测到人脸,所以漏了

Inferencer commented 9 months ago

确保视频尺寸为 1920x1080