Merge branch 'master' into devel

docs/development/01_solution_strategy.rst

@@ -3,7 +3,7 @@ Solution Strategy
 
 Basic decisions for Vision Package:
 
-- Seperation of different tasks into sub-modules (Face Detection, Object tracking, face recognition, mood recognition, age estimation, scene classification, ...)
+- Seperation of different tasks into sub-modules (Face Detection, Object Detection, Object tracking, face recognition, mood recognition, age estimation, scene classification, ...)
 - Highest priority on face detection and face pose estimation. Recognition of people as second priority. All other properties concerning people with lower priority and general object detection with least importance.
 - Face detection using this approach: `Joint Face Detection and Alignment using MTCNNs <https://kpzhang93.github.io/MTCNN_face_detection_alignment/paper/spl.pdf>`_. Good real time performance, other modules to be built on top.
 - Face embeddings using `FaceNet <https://arxiv.org/pdf/1503.03832.pdf>`_. These embeddings can be used for recognition.

docs/development/09_Libraries.rst

@@ -7,6 +7,8 @@ Libraries and external Software
 
   "MTCNN face detection & alignment", "https://github.com/kpzhang93/MTCNN_face_detection_alignment", "MIT license", "Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Neural Networks"
   "Facenet", "https://github.com/davidsandberg/facenet", "MIT License", "Face recognition using Tensorflow"
+  "Object Detection", "https://github.com/pjreddie/darknet", "GNU General Public License", "Object Detection Using YOLO9000"
   "arc42", "http://www.arc42.de/template/", "Creative Commens Attribution license.", "Template for documenting and developing software"
   "DLIB", "https://github.com/davisking/dlib", "Boost Software License", "ML toolkit, mainly used for face detection and facial landmarks"
   "YOLO", "https://github.com/pjreddie/darknet/wiki/YOLO:-Real-Time-Object-Detection", "", "Real-time object recognition system"
+