ZF/OpenFace
1
0
Fork 0
mirror of https://gitcode.com/gh_mirrors/ope/OpenFace.git synced 2026-05-15 03:37:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

- Fixing the issue with gaze not tracking properly in video and landmark modes.

Browse Source 
- Fixing the simscale/simalign bug
This commit is contained in:
Tadas Baltrusaitis
2017-03-08 11:46:50 -05:00
parent a3e66319b5
commit 52c50b4ff3
10 changed files with 226 additions and 256 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
exe/FeatureExtraction/FeatureExtraction.cpp
View File

@@ -308,9 +308,9 @@ int main (int argc, char **argv)
vector<string> output_similarity_align;
vector<string> output_hog_align_files;
double sim_scale = 0.7;
double sim_scale = -1;
int sim_size = 112;
bool grayscale = false;
bool grayscale = false;
bool video_output = false;
bool dynamic = true; // Indicates if a dynamic AU model should be used (dynamic is useful if the video is long enough to include neutral expressions)
int num_hog_rows;
@@ -321,13 +321,13 @@ int main (int argc, char **argv)
bool output_2D_landmarks = true;
bool output_3D_landmarks = true;
bool output_model_params = true;
bool output_pose = true;
bool output_pose = true;
bool output_AUs = true;
bool output_gaze = true;
get_output_feature_params(output_similarity_align, output_hog_align_files, sim_scale, sim_size, grayscale, verbose, dynamic,
output_2D_landmarks, output_3D_landmarks, output_model_params, output_pose, output_AUs, output_gaze, arguments);
// Used for image masking
string tri_loc;
boost::filesystem::path tri_loc_path = boost::filesystem::path("model/tris_68_full.txt");
@@ -391,7 +391,10 @@ int main (int argc, char **argv)
}
// Creating a face analyser that will be used for AU extraction
FaceAnalysis::FaceAnalyser face_analyser(vector<cv::Vec3d>(), 0.7, 112, 112, au_loc, tri_loc);
// Make sure sim_scale is proportional to sim_size if not set
if (sim_scale == -1) sim_scale = sim_size * (0.7 / 112.0);
FaceAnalysis::FaceAnalyser face_analyser(vector<cv::Vec3d>(), sim_scale, sim_size, sim_size, au_loc, tri_loc);
while(!done) // this is not a for loop as we might also be reading from a webcam
{
@@ -593,7 +596,7 @@ int main (int argc, char **argv)
}
if(hog_output_file.is_open())
{
FaceAnalysis::Extract_FHOG_descriptor(hog_descriptor, sim_warped_img, num_hog_rows, num_hog_cols);
face_analyser.GetLatestHOG(hog_descriptor, num_hog_rows, num_hog_cols);
if(visualise_hog && !det_parameters.quiet_mode)
{
@@ -615,13 +618,13 @@ int main (int argc, char **argv)
pose_estimate = LandmarkDetector::GetCorrectedPoseCamera(face_model, fx, fy, cx, cy);
}
if(hog_output_file.is_open())
if (hog_output_file.is_open())
{
output_HOG_frame(&hog_output_file, detection_success, hog_descriptor, num_hog_rows, num_hog_cols);
}
// Write the similarity normalised output
if(!output_similarity_align.empty())
if (!output_similarity_align.empty())
{
if (sim_warped_img.channels() == 3 && grayscale)
@@ -630,18 +633,18 @@ int main (int argc, char **argv)
}
char name[100];
// output the frame number
std::sprintf(name, "frame_det_%06d.bmp", frame_count);
// Filename is based on frame number
std::sprintf(name, "frame_det_%06d.bmp", frame_count + 1);
// Construct the output filename
boost::filesystem::path slash("/");
std::string preferredSlash = slash.make_preferred().string();
string out_file = output_similarity_align[f_n] + preferredSlash + string(name);
bool write_success = imwrite(out_file, sim_warped_img);
if (!write_success)
{
cout << "Could not output similarity aligned image image" << endl;
@@ -1206,6 +1209,7 @@ void get_output_feature_params(vector<string> &output_similarity_aligned, vector
}
// Can process images via directories creating a separate output file per directory
void get_image_input_output_params_feats(vector<vector<string> > &input_image_files, bool& as_video, vector<string> &arguments)
{