RESEARCH

Image & Vision

Investigates Image data fusion techniques that combine image and track data from multiple sensors to achieve improved accuracies and more specific inferences than could be achieved by the use of a single sensor alone. Our aim is to explore the state-of-the-art image processing algorithms for achieving effective data fusion as in:

Face Detection for Classifying Direction Tilted Face

작성자 관리자 날짜 2021-04-08 14:53:23 조회수 52

1. Introduction
    - Using Adaboost training algorithm, object or face detection becomes fast and easy. But it is difficult to detect multi-view
        object by this algorithm. It is a challenging problem in detection algorithm.

2. Main algorithm and principle
    - Face detection for classifying direction of tilted face

1.    Affine transformation of input image

2.    Adaboost classifying about each input images.

3.    Filteration for miss detection

4.    Drawing a face region and classifying face’s condition.

3. Experiment Results
    - Face detection result

 

 

4. Demo
    - Face detection for classifying direction of tilted face

댓글 (0)

등록된 댓글이 없습니다.
작성 권한이 없습니다.