While capturing video thru OpenCV, I'm getting following. [closed]
D:\Ram\Projects\Temp and Samples\dlib-18.18\examples\build\Debug>makeupapp_ex.exe shape_predictor_68_face_landmarks.dat
Error detected at line 36. Error detected in file d:\ram\projects\temp and samples\dlib-18.18\dlib\opencv/cv_image.h. Error detected in function __thiscall dlib::cv_image::cv_image(const class cv::Mat).
Failing expression was img.depth() == cv::DataType::basic_pixel_type>::depth && img.channels() == pixel_traits::num. The pixel type you gave doesn't match pixel used by the open cv Mat object. img.depth(): 0 img.cv::DataType::basic_pixel_type>::depth: 0 img.channels(): 1 img.pixel_traits::num: 3
I am using Dlib & OpenCV for facial landmarking
The error is generated in dlib and thus this question is not related to OpenCV but rather to a wrapped version. Since this is not the core buisness of this forum, please post your question at the correct place.
let me clarify: you need to convert to grayscale, before passing it to dlib's shape prediction.