I am a researcher at XEROX’s Research Center Europe in Grenoble France. Here in Grenoble, we are researching methods for automatically assessing the quality and aesthetics of photographs (please see link here). We hope to use these assessments to create photo albums, catalogs, or other similar products.
We believe that the information found on www.dpchallenge.com, in particular the average scores associated with a specific photograph for a given challenge, can be very useful for helping us understand how people judge an image. We would like to collect and use this information in our research in two ways:
First, we would like to collect photographs submitted to a challenge and process them together with their associated data (such as their average scores) in order to train and test image assessment algorithms.
Second, we would like to include these photographs and their associated data in a data set that we are trying to assemble for use in benchmarking image analysis algorithms. Once fully assembled, we would like to distribute the dataset to members in the image analysis research community for use in their own research.
Such a benchmarking tool would help our research community to assess the overall progress of the different image analysis methods that are being developed.
We would like to include the photos you have submitted to www.dpchallenge.com in this dataset. As such, I was wondering if you would be so kind as to give your consent to such terms of use.
:) Not surprised to see a post like this. I've mentioned the possibility for this in a few posts here. In fact, I'd be surprised if Topaz Labs doesn't have something already in the works.
It probably wouldn't work very well, though. It would be extremely difficult to factor in judgement calls by voters when they consider how well something meets the challenge or not.
An outstanding image in the wrong challenge would not score well at all here.
edit to add: If they limited their data collection to free study challenges, then that would avoid the DNMC problem.
Oh, I contacted SC before I saw this thread. I got the same message, and it made me uncomfortable. Xerox is a legitimate corp. with a good reputation, that does a lot of research, but I couldn't help wondering if this was SPAM or phishing. It was very well written. How are the rest of you responding?
A very quick Google search on "Naila Murray Xerox" confirms that her name does in fact turn up on publications as a Xerox researcher. The use seems legit. I have not received the message (no real surprise there, I haven't entered much in quite a while) but if I did, I believe I would grant the rights for the use.
I didn't get this..obviously, lol...but the thing that stood out for me was the distribution of the images to others...but if they are using one FS, guess that's only one photo for each person. Never mind...
I am a researcher at XEROX’s Research Center Europe in Grenoble France. Here in Grenoble, we are researching methods for automatically assessing the quality and aesthetics of photographs (please see link here). We hope to use these assessments to create photo albums, catalogs, or other similar products.
We believe that the information found on www.dpchallenge.com, in particular the average scores associated with a specific photograph for a given challenge, can be very useful for helping us understand how people judge an image. We would like to collect and use this information in our research in two ways:
First, we would like to collect photographs submitted to a challenge and process them together with their associated data (such as their average scores) in order to train and test image assessment algorithms.
Second, we would like to include these photographs and their associated data in a data set that we are trying to assemble for use in benchmarking image analysis algorithms. Once fully assembled, we would like to distribute the dataset to members in the image analysis research community for use in their own research.
Such a benchmarking tool would help our research community to assess the overall progress of the different image analysis methods that are being developed.
We would like to include the photos you have submitted to www.dpchallenge.com in this dataset. As such, I was wondering if you would be so kind as to give your consent to such terms of use.
Thanks very much indeed!
Naila Murray
As I said, I received the same message but there was not link where it says "Please see link here". How can we find out which image they are referring to?
Curious that the person chose to contact individual members rather than the admins or SC. Surely it would have been easier and less suspect to get official backing and call for volunteers?
Curious that the person chose to contact individual members rather than the admins or SC. Surely it would have been easier and less suspect to get official backing and call for volunteers?
I suspect that they may have some criteria that they are applying to select images. They definitely need the photog's permission, but of course they do not need the site's permission. The score is public information. Of course it's still not a bad idea for them to contact the Site Administration. Of course, we don't know that they haven't.
Based on what I'm reading here, especially the response to Bob's inquiry, I'm inclined to give my permission. But I want to retain copyright ownership of my images.
That may or may not be. My point was to for US to THINK about what the possible uses might be.
Oh, absolutely. Bottom line, each of those contacted will want to think about if they are comfortable "supporting" that research via the use of their images.
On the separate question of retention of copyright, I think all they are asking for is limited use.
