diff options
| author | Michael Welter <michael@welter-4d.de> | 2022-05-12 19:54:57 +0200 |
|---|---|---|
| committer | Michael Welter <michael@welter-4d.de> | 2022-05-19 20:46:50 +0200 |
| commit | 4d6e80a6c585f6929f6bf4bf319932732160ce84 (patch) | |
| tree | ffbef0e4a2625afccd9b3b1c3da682f54d1c32a1 /proto-wine/images | |
| parent | 25af3db8ba560d86f78aa4bfd10039588b7bab82 (diff) | |
tracker/nn: Add support for recurrent model and further tweaks
Regarding tweaks:
* EWA smoothing of head ROI. Smoothing strength is a UI setting.
* Adjustible zoom into the detected face. The predicted ROI is scaled by a factor the user can set. There is a sweet spot somewhere near 1.
* Adjustible number of threads
* The ROI is no longer taken as model output directly. This was actually not needed. Perhaps as auxiliary training objective for the network. But the tracker implementation now just uses a square area around the head center according to the predicted head size.
* Add comment and debug notification on face ROI model output
Diffstat (limited to 'proto-wine/images')
0 files changed, 0 insertions, 0 deletions