From: shastri@umail.ucsb.edu
Subject: still missing...
Date: March 4, 2004 6:51:01 PM PST
To: manj@ece.ucsb.edu
Cc: surreal@umail.ucsb.edu
Dear Professor Manjunath:
It looks as if there are still files missing for Group6's attacks.
In the file, s_GetFreqSamples.m, there is a function call "s_GetFreqSampleSize(
img )" on line 42.
s_GetFreqSamples.m notes the use of the function "s_GetFreqSampleSize" in the
comments at the beginning, but unfortunately the function is not included.
Sincerely,
Shobhna Shastri (Group 5)
From: yonester@hotmail.com
Subject: Re: problem with attack03Script.m
Date: March 5, 2004 2:04:00 AM PST
To: manj@ece.ucsb.edu
Sorry about that, this was just a misunderstanding of the instructions.
We'll submit the new code ASAP.
Yony
----- Original Message -----
From: "B. S. Manjunath" <manj@ece.ucsb.edu>
To: "Gemste" <gemste@yahoo.com>
Cc: "Yony K." <yonester@hotmail.com>; "Manjunath B. S." <manj@ece.ucsb.edu>
Sent: Thursday, March 04, 2004 12:35 PM
Subject: Re: problem with attack03Script.m
Group 3: please respond and copy to me. I will also post these
correspondences on a web page.
The input should be an image containing hidden data. Since we are
requiring that the scheme be robust, at the minimum, to JPEG
compression, the attack scheme should accept a JPG image and write out
a JPG image at 50% quality.
If there are deviations, they should be justified.
-Manjunath
On Mar 4, 2004, at 12:18 PM, Gemste wrote:
Their attack requires a a bitmap file and writes a
bitmap file for the output.
line3 - im = imread('178 - HwkBoxes.bmp');
line41- imwrite(im.204attacked.bmp', 'BMP');
Should I adjust my code to write a jpg and a BMP file?
Or should i have that attack write out as a jpg?
-David li
group 5
From: surreal@umail.ucsb.edu
Subject: Re: Attack Schemes
Date: March 4, 2004 1:06:43 PM PST
To: manj@ECE.UCSB.EDU
Dr. Manjunath,
I have resolved serveral dependencies in the matlab code we posted earlier. The
three attached files need to be present to run the attacks.
Sorry. I'm not used to source based releases.
Thanks,
Terrence Cole
(group 6)
Quoting "B. S. Manjunath" <manj@ece.ucsb.edu>:
done. please check and make sure that the web site is properly updated.
-Manjunath
On Mar 4, 2004, at 9:59 AM, Terrence Cole wrote:
<smile.jpg>
--
Terrence Cole
surreal@umail.ucsb.edu
____________________________________________________________________________________________
From: hsi@engineering.ucsb.edu
Subject: file missing?
Date: March 4, 2004 12:54:19 PM PST
To: manj@ece.ucsb.edu
Dear Prof:
It seems like there is a file missing for Group6's second attack.
In the file, attackG06_swap_q.m, there is a function call "s_ResizeUpToEight(
img )" on line 5; however, the function is not provided. Thanks.
Sincerely,
Sean Wu of Group 5
______________________________________________________________________________________________
03/04 1238PM BSM
Group 3: please respond (see below)and copy to me. I will also post these correspondences on a web page.
The input should be an image containing hidden data. Since we are requiring that the scheme be robust, at the minimum, to JPEG compression, the attack scheme should accept a JPG image and write out a JPG image at 50% quality.
If there are deviations, they should be justified.
-Manjunath
On Mar 4, 2004, at 12:18 PM, Gemste wrote:
Their attack requires a a bitmap file and writes a
bitmap file for the output.
line3 - im = imread('178 - HwkBoxes.bmp');
line41- imwrite(im2, 'attacked.bmp', 'BMP');
Should I adjust my code to write a jpg and a BMP file?
Or should i have that attack write out as a jpg?
-David li
group 5
=====