search

audeering / opensmile

The Munich Open-Source Large-Scale Multimedia Feature Extractor
https://audeering.github.io/opensmile/
Other
553 stars 74 forks source link

Extrcting features with opencv #39

Open nidhijoshi123 opened 2 years ago

nidhijoshi123 commented 2 years ago

Thanks for the software!

I am getting the following error when I provide the audio visual input:

(ERR) [0] commandlineParser: parse: unknown option '-P' on commandline! (ERR) [1] cComponentManager: cannot add component (instname='openCVSource' type='cOpenCVSource'): unknown component type!! (ERR) [1] cComponentManager: error during addComponent (returnVal=-1)!

Can you please help ? Do I need to externally download some models ?

chausner-audeering commented 2 years ago

(ERR) [1] cComponentManager: cannot add component (instname='openCVSource' type='cOpenCVSource'): unknown component type!!

This indicates that openSMILE was not built with OpenCV. The pre-built binaries made available via GitHub do not include OpenCV, you will have to build it yourself. See https://audeering.github.io/opensmile/get-started.html#compiling-on-linux-mac-with-portaudio-and-opencv-support and https://audeering.github.io/opensmile/get-started.html#compiling-on-windows-with-portaudio-and-opencv-support in the documentation.

chausner-audeering commented 2 years ago

Do I need to externally download some models ?

openSMILE does not come with any models for audio-visual analysis and it is also not possible to run models as part of openSMILE. You can only use openSMILE to extract features. Note that visual feature extraction in openSMILE is extremely limited, as the focus for openSMILE has always been on audio feature extraction.