Academic Andrea Tagliasacchi defended the award's value in academic settings.
I don't care much about the CVPR Best Paper Award criteria.
Usually it means nothing, as least much less than Test of Time award. It never mattered before, it will never matter in the future. I will change my mind, when the "Best Paper" will also get test of time award.
OK, it seems I have to change my mind.
I don't care much about the CVPR Best Paper Award criteria.
Usually it means nothing, as least much less than Test of Time award. It never mattered before, it will never matter in the future. I will change my mind, when the "Best Paper" will also get test of time award.
@ducha_aiki I agree with the sentiment, but I’m pretty sure ResNet just won test of time, and also won CVPR 2015 best paper award??
@ducha_aiki In academia they do carry weight. So following your argument, if in industry they don’t matter… then they should just be restricted to pure academia.
I don't care much about the CVPR Best Paper Award criteria.
Usually it means nothing, as least much less than Test of Time award. It never mattered before, it will never matter in the future. I will change my mind, when the "Best Paper" will also get test of time award.
@taiyasaki I am genuinely wondering which ones exactly. You also probably should downweight my opinion, as I never got any oral presentation, so awards are definitely out of the scope.
@ducha_aiki Absolutely.
In fact it has been challenging at times to tell student to ignore the hype (as it is compute restricted) and focus on producing high quality technical work.
@ducha_aiki Seems unlikely (there seems to be a single CVPR paper that won both awards). But anyways, if a paper wins test of time, they will have had impact, so it shouldn’t matter if they released source code originally or not.
@JonathonLuiten You are wrong
https://cvpr2015.thecvf.com/awards.php
@ducha_aiki I stand corrected, there are two such papers.
@SattlerTorsten What are those?
@ducha_aiki @JonathonLuiten I think it was 2016
@ducha_aiki you are probably asking for non-existent capabilities in humans. They (we) ultimately decide Best Paper vs just a poster vs reject
@ducha_aiki Resnet and “Object Class Recognition by Unsupervised Scale-Invariant Learning”
@ducha_aiki I got the year wrong (from memory) but the vibes right! Was 2016 (as pointed out below)
@KadirYilmaz_CV @JonathonLuiten You are right about 2016
https://tc.computer.org/tcpami/awards/cvpr-paper-awards/
@JonathonLuiten @ducha_aiki But resnet had code and pretrained models available ;)
Academic Andrea Tagliasacchi defended the award's value in academic settings.
I don't care much about the CVPR Best Paper Award criteria.
Usually it means nothing, as least much less than Test of Time award. It never mattered before, it will never matter in the future. I will change my mind, when the "Best Paper" will also get test of time award.