1. Efficient QR Code Beautification With High Quality Visual Content
Source: IEEE Transactions on Multimedia 17.9 (2015):
Authors: Shih-Syun Lin, Min-Chun Hu, Chien-Han Lee,and Tong- Yee Lee
Speaker: Peng-Cheng Huang
Date: 12/2/2018

2. Outline Introduction Related Work Proposed Scheme Experimental Results
Conclusions

3. Introduction(1/3) msn.iecs.fcu.edu.tw
The stander QR code usually looks like random black-and–white patterns. You can not find anything out even if you stare it all day. Actually , this QR code contain the URL of our laboratory. But if we add some aesthetic elements to make the original QR code looks like this one, and this one . I think these QR codes are much more attractive than original one.
msn.iecs.fcu.edu.tw

4. Intrudoction(2/3) n-bits RS code.
This is a n bits RS code, including m bits for message bits, p bits for padding bits, and t bits for parity bits. Parity bit used for detecting and correcting errors during scanning the QR code.
The function pattern used for improving the reading performance.
n-bits RS code.

5. Intrudoction(3/3) Two kinds of variations to beautify the QR code
in the past few years , many researcher have tried to beautify the QR code by adding aesthetic elements or visual content . The are two kind of methods to beautify the QR code : changing the shape/color of modules and embedding a picture into the QR code. Here are some examples.
Two kinds of variations to beautify the QR code

6. Related Work: Reed-Solomon Code(1/3)
Background: Reed-Solomon encoded messages can be XOR'ed
Reed-Solomon coding
m + p = k t
Where demotes the bit stream exclusive-or(XOR) operation.
Reed-Solomon coding k1 t1
Reed-Solomon coding k2 t2
Reed-Solomon coding
k1 ⊕ k2
t1 ⊕ t2
 ⊕ means XOR

7. Related Work – Russ Cox’s Method(2/3)
Controls bits(16 bits)=message(2bits)+pad(14bits)
Parity bits:16 bits
QR code data:
Image data:
Basis set of the entire vector space of valid Reed-Solomon blocks.

8. Related Work – Russ Cox’s Method(3/3)
QR code data:
Image data:
Controls bits :16 bits
Parity bits:16 bits

9. Proposed Scheme(1/5) The flowchart of the proposed QR code beautifier
we propose a two-stage approach to generate QR code with high quality visual content. In the first stage, a baseline QR code with reliable decodability but poor visual quality is first synthesized based on the Gauss-Jordane limination procedure. In the second stage, a rendering mechanism is designed to improve the visual quality while avoiding affecting the decodability of the QRcode. The experimental results show that the proposed method substantially enhances the appearance of the QR code and the processing complexity is near real-time.
The flowchart of the proposed QR code beautifier

10. Proposed Scheme – Baseline QR Code Synthesis(2/5)
Example: three 5-bits RS codes with k = 3 and t = 2 :
RS2 = RS1  ⊕ RS2
K4,K5,k6 can be generated by applying XOR operations on the RS blocks of the basis set

11. Proposed Scheme -- Baseline QR Code Synthesis(3/5)
(a) Illustration of how an RS decodable bit stream is placed onto
a 2D QR code. (b) The new bit stream obtained by changing positions of
control bits and the corresponding 2D QR code. Border & Central

12. Proposed Scheme – QR Code Rendering(4/5)

13. Proposed Scheme – QR Code Rendering(5/5)
where

14. Results:Visual quality and time complexity(1/8)
each of them has limitations in terms of the size constraint of the embedded picture, visual quality of the generated QR code, or computational complexity
Visualead
halftone
Lin[9]
This paper

15. The influence of α value(2/8)

16. Correctness of QR code decoding(3/8)

17. The influence of image size and QR code size(4/8)

18. The influence of ρx and ρy(5/8)

19. The influence of error correction level(6/8)

20. The influence of message length(7/8)

21. The influence of lighting condition and viewing angle(8/8)

22. Conclusions The experimental results show that：
the proposed method outperforms the existing works considering the appearance of the QR code
the processing complexity is near real-time.
In this work, we propose an efficient two-stage approach to generate QR codes with high quality visual content. A decodable baseline QR code with poor visual quality is first synthesized based on the Gauss-Jordan elimination procedure and then a rendering mechanism is designed to improve the visual quality while avoid affecting the decodability of the QR code.