[IPOL discuss] Ipol and Dataset

[Jean-Michel Morel]

I  am quite in favor of this. I'll discuss it with Nicolas.
One of the things we really need badly are stereo pairs of multiple 
images with ground truth attached to one of the images. We found no 
other way  so far as to use image synthesis to create such ground truth.
(We are also aware of the Strecha database 
http://cvlab.epfl.ch/~strecha/multiview/rawMVS.html)

Best,
Jean-Michel

Daniel Kondermann a écrit :
> We do have a fully featured dataset/benchmark webseite up and running at
> our lab including user management and expiring download-links. I would
> suggest to cross-link both sites and upload such data on our site.
> http://hci.iwr.uni-heidelberg.de/Benchmarks/
> 
> The site is a long-term strategy element of the HCI and is continuously
> being further developed. In case you have any feature-request, we can
> implement them.
> 
> Best,
> Daniel
> 
> Am 26.05.2011 21:31, schrieb Pierre Moulon:
>> Hi,
>>
>> Following the recent post about 3D point cloud data.
>>
>> Does it have sense to open a "Dataset section" in IPOL rather than put it in
>> the algo section ?
>>
>> http://www.ipol.im/pub/algo/d_point_cloud_data/
>>
>> Other dataset could be added like the one from Guoshen and Morel that they
>> have done for the ASIFT/ SIFT affine invariance.
>> Aka : Dataset: An image dataset for systematic evaluation of robustness to
>> absolute and transition tilt of the image matching algorithms is available :
>> zip tar/gz.
>> >From here http://www.ipol.im/pub/algo/my_affine_sift/.
>>
>> Regards,
>> Pierre M
>>
>>
>>
>>
>> _______________________________________________
>> discuss mailing list
>> discuss at list.ipol.im
>> http://tools.ipol.im/mailman/listinfo/discuss
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> discuss mailing list
> discuss at list.ipol.im
> http://tools.ipol.im/mailman/listinfo/discuss