432412 2021-02-02 10:34:36 +0000 Training Face recognition for blur and pixellation 2025-12-25 10:54:57 +0000 1 1 2 Applications digikam Faces-Detection unspecified macOS (DMG) macOS RESOLVED FIXED NOR wishlist --- 0 paulpuaux digikam-bugs-null caulier.gilles metzpinguin 9.0.0 0 oldest_to_newest 1999580 0 paulpuaux 2021-02-02 10:34:36 +0000 SUMMARY Offering visual UX method to exclude from face recognition faces too blurred or too pixelated, using Deep Learning STEPS TO REPRODUCE 1. Adding a button/box with blur/pixel icon 2. In popup window, show on the left an illustrated blur slider to validate a requested sharpness index (i.e. above a blur factor of x) 3. show on the right an illustrated pixel slider to validate a requested precision index (i.e. above a pixel factor of x) OBSERVED RESULT Exclude from listing low quality / poorly performing recognized faces EXPECTED RESULT Get a more consistent visual face album, less blurred/pixelated results, while maintaining fast accurate recognition. SOFTWARE/OS VERSIONS Windows: macOS: Linux/KDE Plasma: (available in About System) KDE Plasma Version: KDE Frameworks Version: Qt Version: ADDITIONAL INFORMATION 1999583 1 metzpinguin 2021-02-02 10:55:32 +0000 *** This bug has been marked as a duplicate of bug 425263 *** 2477131 2 caulier.gilles 2025-12-15 14:43:58 +0000 Fixed with bug 425263