794 karma · joined April 27, 2011
xavier@whirlscape.com @wxswxs
[ my public key: https://keybase.io/wxswxs; my proof: https://keybase.io/wxswxs/sigs/fhibuu3BrSc7k3fLOwDrKBHC1A5GGzzsoPACnipuWJ4 ]
http://itunes.apple.com/us/app/vital-signs-camera-philips/id...
http://people.csail.mit.edu/celiu/motionmag/motionmag.html
This other algorithm amplifies only motion, and the algorithm is less elegantly simple, however its results are very interesting and fun to watch as well. Make sure to watch the video at the bottom of the page
http://people.csail.mit.edu/celiu/motionmag/VIDEO_siggraph_2...
I see it as more useful in that it eliminates the issue where you take a great photo and then realize that the focus was off. It's a shame the resolution is so low, but I think this is a very interesting first commercial foray into the area.
A glossary of the intelligence jargon that they use, written in an entertaining style.
Also, with regard to your comment about making pixels look different depending on the angle, there's an alternate approach to this using stacked LCDs that's getting interesting results.
From the diagram, it looks like they're projecting onto a glass screen in front of the object, and it appears to be superimposed on the object using the same approach as the Pepper's ghost box [1].
Of course I may be mistaken. This is just an educated guess.
And regardless, this sounds like a really cool technology.
[1] http://en.wikipedia.org/wiki/Peppers_ghost
EDIT: From the Google Translate of the Japanese press-release, this looks correct, they are using a half-silvered mirror so you can see the egg-through the mirror and also the projection reflected on the mirror
" Hitachi, respond to these needs, aiming to represent the object close to the actual images from nature, using a half mirror for combining multiple lenses and translucent, the image information captured in real-space projection of 24 units has enabled it to be displayed".
EDIT: This definitely works for Safari, not 100% sure if it does for Chrome after all.
http://blog.thenounproject.com/post/7310229014/how-to-proper...
"When using a CC Attribution work created by us, please attribute it to The Noun Project and include a link whenever possible." (the post goes into more detail for CC-BY in general as well)
The front-page is big enough that I can handle there being one post for the original raw-data, and a subsequent post for the Ars analysis.
Your other point about "cheerleading" is well taken, It is easy to get caught up in the David vs. Goliath sentiment.
Shifted 1 week: gpa calculator
Shifted 2 weeks: final grades
Shifted 3 weeks: academic suspension
Shifted 4 weeks: academic dismissal
http://correlate.googlelabs.com/search?e=exam+schedule&t...
http://correlate.googlelabs.com/search?e=exam+schedule&t...
http://ngrams.googlelabs.com/graph?content=love+makes&ye... http://ngrams.googlelabs.com/graph?content=love+with&yea... http://ngrams.googlelabs.com/graph?content=love+me&year_... http://ngrams.googlelabs.com/graph?content=love+you&year... http://ngrams.googlelabs.com/graph?content=you+love&year...
etc.
If anything, though, I feel like this may be due to a changing in the makeup of the corpus in recent years. Many queries with common words seem to exhibit the same spike after around 2000 (try "hate", "fly", "ten", "blue", etc.) perhaps the size of the average vocabulary is smaller in the newer corpus documents, in which case the overall proportion of common words would be expected to increase.
For older content, they focus on what's available in libraries so perhaps old content that's survived in libraries typically has a richer vocabulary.
Or maybe we just love more.
Side note, this may disappear after the course is over. For anyone who is interested in going through this on their own later, might I recommend
wget -rkp -np --cut-dirs=1 http://www.ee.columbia.edu/~dpwe/e4896/
(Warning, this includes the lecture videos so will be multi-GB)