face detection and recognition for distributed systems meng lin and ermin hodžić 1

16
Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Upload: brett-carr

Post on 29-Dec-2015

216 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Face Detection And Recognition ForDistributed SystemsMeng Lin and Ermin Hodžić

1

Page 2: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Motivation

• Security systems•Digital cameras, adjustments• Social networks•Marketing

2

Page 3: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Face Detection

• Locating and extracting faces in images•Rectangles as output

3

Page 4: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Face Recognition

• Identifying person on image• Finding closest match

4

Page 5: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

MapReduce model

•Parallel model• Framework for distributed systems•Hadoop local filesystem•Amazon Elastic MapReduce

5

Page 6: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Solution: Face Detection

6

Page 7: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Solution: Face Detection

7

Page 8: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Solution: Face Detection

•Partitions based on face scale

8

Page 9: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Solution: Face Recognition

•Distribution of recognizers•Recognizing in parallel•Reduce the most confident result

9

Page 10: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Solution: Face Recognition

10

Page 11: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Hadoop Image Handling

• Images as text•Collection of images in a big file•Utilize Hadoop default input format

11

Page 12: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Hadoop Image Handling

12

Page 13: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Evaluation

• Segmented Face Detection

•Raw OpenCV Face Detection

13

Page 14: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Evaluation

• Face recognition

•More nodes = slower• Input and data transfer overhead• Jobs computationally cheap

14

Page 15: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

Conclusion

• Easily scalable parallel mode•Generalized framework• OpenCV just one sample tool

•A lot of communication• Low utilization of processors

15

Page 16: Face Detection And Recognition For Distributed Systems Meng Lin and Ermin Hodžić 1

16