search

thuiar / MMSA-FET

A Tool for extracting multimodal features from videos.
GNU General Public License v3.0
141 stars 20 forks source link

更换自己的数据集后MMSA-FET无法正常提取特征 #22

Open Chi-Tian opened 1 year ago

Chi-Tian commented 1 year ago

我按照提示，按照论文方式裁剪视频，然后按照SIMS的文件结构构建，更换自己的数据集路径以后，无法用MMSA-FET提取多模态特征，用单个视频运行run_single( )可以提取，但是换成数据集，使用run_dataset( )则无法正常提取，所有视频片段全部跳过，会报错“ MMSA-FET - Error occurred when extracting features for video video_0001 clip 1 MMSA-FET - Ignore error and continue, see the log for details. MMSA-FET - Error occurred when extracting features for video video_0001 clip 2 MMSA-FET - Ignore error and continue, see the log for details” 请问这是哪里的问题，麻烦您回复.

FlameSky-S commented 1 year ago

视频特征提取使用的是OpenFace吗？是否有更详细的报错信息？