GSoC/2019/StatusReports/ThanhTrungDinh: Difference between revisions

Revision as of 13:41, 23 June 2019

digiKam AI Face Recognition with OpenCV DNN module

digiKam is KDE desktop application for photos management. For a long time, digiKam team has put a lot of efforts to develop face engine, a feature allowing to scan user photos and suggest face tags automatically basing on pre-tagged faces by users. However, that functionality is currently deactivated in digiKam, as it is slow while not adequately accurate. Thus, this project aims to improve the performance and accuracy of facial recognition in digiKam by exploiting state-of-the-art neural network models in AI and machine learning, combining with highly-optimized OpenCV DNN module.

The project includes 2 main parts:

Improve face recognition: implementation with OpenCV DNN module
- reduce processing time while keeping high accuracy
- classify unknown faces into classes of similar faces
Improve face detection: implementation to be investigated
- detect faces across various scales (e.g. big, small, etc.), with occlusion (e.g. sunglasses, scarf, mask etc.), with different orientations (e.g. up, down, left, right, side-face etc.)

Mentors : Maik Qualmann, Gilles Caulier, Stefan Müller

Work report

Bonding period (May 6 to May 27)

Coding period : Phase one (May 28 to June 23)

Important Links

Proposal Link

Project Proposal

Git dev branch

gsoc19-face-recognition

Contribution

Contacts

Email: [email protected]

Github: TrungDinhT

@@ Line 1: / Line 1: @@
-= digiKam AI Face Recognition with OpenCV DNN module =
+== digiKam AI Face Recognition with OpenCV DNN module ==
+digiKam is KDE desktop application for photos management. For a long time, digiKam team has put a lot of efforts to develop face engine, a feature allowing to scan user photos and suggest face tags automatically basing on pre-tagged faces by users. However, that functionality is currently deactivated in digiKam, as it is slow while not adequately accurate. Thus, this project aims to improve the performance and accuracy of facial recognition in digiKam by exploiting state-of-the-art neural network models in AI and machine learning, combining with highly-optimized OpenCV DNN module.
+The project includes 2 main parts:
+* '''Improve face recognition''': ''implementation with OpenCV DNN module''
+** reduce processing time while keeping high accuracy
+** classify unknown faces into classes of similar faces
+* '''Improve face detection''': ''implementation to be investigated''
+** detect faces across various scales (e.g. big, small, etc.), with occlusion (e.g. sunglasses, scarf, mask etc.), with different orientations (e.g. up, down, left, right, side-face etc.)
-digiKam is a well-known desktop application for photos management. In digiKam, tags on photos are strongly supported for the sake of providing users with a natural workflow of searching and arranging photos in their collections. Since many of our photos contain faces, face tag has apparently emerged as an essential property for any photos management software. Being aware of that, digiKam team has put a lot of efforts to develop face engine, which scan scan photos and suggest face tags automatically basing on pre-tagged photos by users. However, that functionality is currently deactivated in digiKam, as it is slow while not adequately accurate. Thus, this project aims to improve the performance and accuracy of facial recognition in digiKam, in order to bring this wonderful functionality back to users in a very soon release.
 '''Mentors''' : Maik Qualmann, Gilles Caulier, Stefan Müller
-== Project Goals ==
-*Implement DNN based approach and unit tests with OpenCV DNN module
-*Complete integration tests on computational and accuracy benchmark for face engine
-*Study performance metrics and decide which algorithm and which kind of neural network architecture to use for facial recognition in digiKam
-*(Optionally) Implement facial detection with OpenCV DNN module to replace current method using Haar cascade algorithm
 == Work report ==