Subscribe to Computing Intelligence

Tuesday, August 20, 2013

Proverbs of Computer Vision and Robotics

I've put up what I think is a fun post at my Computing Intelligence blog:

Proverbs of Computer Vision and Robotics

Since this is the first time I've updated that blog in several years, I thought it might be worth posting a link here, too (not that I have much activity here, but it is at least a little more active!).

Wednesday, October 31, 2012

Halloween Accessories: How to Make a Buckler

My Halloween costume this year never quite came to fruition, but over the last couple weeks I nevertheless managed to create a number of successful accessories.  Here, then, is part one of my guide to making a passable buckler, battleaxe, and sword.

Part I: The Buckler
A buckler is a small shield, generally metal and 15-45cm in diameter, gripped in the fist.  I honestly used to think that a buckler was actually a small shield which straps (or buckles, hence the name) to the forearm, but Wikipedia leads me to believe that this is actually an inaccurate view arising from Dungeons and Dragons.  A buckler has a number of advantages as a Halloween costume prop over a larger shield: it takes less material to make, and it is also a lot easier to carry around at a party.  The easiest way to make a buckler is to go to a thrift store which carries dishes and find a round metal bowl (in Toronto, the two best places to look would likely be Honest Eds or a Value Village).
The suitably shaped bowl I used to make my buckler.
Turing the bowl into a buckler is then simply a matter of attaching a handle to carry it.  If one is looking to put more distinction between one's buckler and simply a bowl with a handle, an additional option would be to attach a central spike (bucklers were not only defensive, but could also be used as a punching weapon and were occasionally adorned with spikes or sharpened edges for that purpose).  Although the traditional buckler handle was in the direct centre, I found it easier to control by placing a handle slightly off centre with a corresponding forearm strap.  In order to get away with only a central handle it would have to be anchored fairly strongly to prevent the buckler from twisting in one's grip; using a strap and a handle is more forgiving.
Buckler handle and forearm strap.
I wasn't particularly careful with the appearance of the handle, as the underside of the buckler would generally be hidden.  In order to give it some rigidity, I started with a wooden chopstick.  I then took an old sock which I used to protect my hand when tousling with the cat (not to worry, I have several pairless socks which have been donated to this cause, so she will not mourn the loss of this one) and used part of it to wrap a layer of cloth around the chopstick.  This provided both extra thickness to make the handle easier to hold and "tabs" which I could tape to the surface of the bowl to hold the handle in place.  When using tape to secure anything which will need to support weight, it is best to align several strips in alternating perpendicular patterns.  After I attached the handle I used the remains of the sock to place a pad on the inner surface of the bowl to prevent my knuckles from rubbing against the bare metal.

My initial strap was a thick rubber band which was also held on with tape, but it ended up snapping in what we believe to be a cat-related incident.  Since I had the hot glue-gun out anyway for the sword, I decided to use it to secure a new strap made from two lengths of string wound together.

That wraps up the simplest of the three accessories.  Altogether it cost me $3.50 for the bowl, an old sock, some string, a random chopstick I found in our kitchen drawer, some duct tape, and a couple dollops of hot glue.  Here is an image of the buckler being held up to parry a strike:

Buckler parrying an axe.
The axe will be described in the next part.

Friday, October 19, 2012

Master of Vision

As those who follow this blog may have noticed, I again entered a bubble of silence since mid-summer.  Rather than another bout of writer's block, however, I was instead under deadline pressure to finish my Master's thesis, and time spent writing other things seemed wrong.  I am happy to report that on October 5th I defended my thesis, which was accepted without revision.  For anyone who is interested in plowing through it, a finalized copy of my thesis can be downloaded from here.  Now that my thesis is finished, it is time for me to get busy on my PhD and all of the small items (like marking) which got pushed to the back-burner leading up to the defense.  I am hoping that one of those small items which I will now have more time for is blogging regularly again.

Tuesday, July 10, 2012

Excellent Climate Change Allegory

These past few weeks have involved long hours sitting at my computer trying to hammer out my thesis and ignore the uncomfortable sensation of sweat sticking me to my chair or trickling down my side. Thankfully, I do not suffer from hyperhidrosis, but it has simply been incredibly hot and humid.  In fact, this summer has brought on another record set of heat waves, which one can only hope will finally galvanize political environmental action.  Of course, this hope is likely in vain.  Media Matters published an excellent article on the coverage of the recent heat wave, which is worth reading:
In December 2008 the Washington Post reported that AT&T and DuPont planned to lay off a combined 14,500 employees. The lead of the story said: "Need more proof that the recession is real? An onslaught of grim unemployment and layoff reports yesterday should dispel any lingering doubts."
Was the recession the only force behind these job cuts? No. Other variables would be needed to explain why the layoffs were hitting these specific companies, at this time, and at this scale. But the recession was the obvious background condition, the broader context that could not go unmentioned in a proper news report on the layoffs, and there was no hand-wringing about drawing the connection. The article didn't caution that "No single bankruptcy or job cut can be definitively blamed on the recession." No one waited for a computer model to precisely sort the causes of these layoffs. No one tracked down a contrarian to point out that layoffs happened long before the recession and that, in fact, such-and-such a company somewhere is hiring.
Which brings me to the massive heat wave that we're now emerging from. Scientific observation and analysis have established that human-induced climate change makes extreme heat events more common. But when heat waves hit, many reporters hesitate to mention climate change without appending disclaimers of the sort that you don't see on other beats.
Go read the rest of the article.  As with all allegories, it is obviously not a perfect fit, but it does an excellent job of showcasing the frustrating manner in which environmental issues, particularly climate change, is treated by the media.

Friday, June 22, 2012

Some Friday Humour

Given that it is a Friday and some sort of ridiculous construction is taking place just below my window which makes concentrating on work rather difficult, I thought I would post something lighthearted.  Here are three videos of Henri, a cat undergoing an existential crisis.  I think the videos get progressively better (both in film quality and content), but it is nevertheless worth watching them in order to see the development of poor Henri's nihilism.





Sunday, June 17, 2012

What does Stephen Harper think public scientists are for?


Stephen Harper and the Conservative Party of Canada have never provided a convincing illusion of being good for either science or the environment, but they nevertheless managed to wheedle enough support to squeeze out a majority government*.  Somehow, a slim majority has put it into Harper's head that this now means that his party doesn't just lead the Government of Canada, but rather that they are the Government of Canada (or, rather, that the Government of Canada shall now henceforth be renamed the Harper Government).  Taking this
conflation and running with it, Harper has consistently taken the view that government scientists must get approval before speaking with the press.  The latest example of this policy is a letter sent to Parks Canada employees informing them that it is their "duty" to support the Harper government.  The bizarre logic behind this policy seems to be that Harper views Canadian scientists as employees of the government, and since he views the Government of Canada and the Harper Government as synonymous entities, all Canadian scientists (and all other civil servants, by the same logic) are now expected to toe the Conservative line.

Which leads me to conclude that Harper must not understand what public science is for.  After all, supporting party policy is the job of politicians and pundits, not scientists.  It is this fundamental misunderstanding, then, which leads to the short-sighted axing of huge swathes of Canadian science.  For example, the government claims that halting funding for the Experimental Lakes Area makes sense because it no longer fits with Ottawa's mandate.  Such a line of reasoning actually does make sense if one believes that the job of government scientists is to support government policy, since the current government doesn't actually care about fresh water preservation or protection.

Civil science in actuality is meant to service the people of the country, not the policy of the government.  Of course, this is an old struggle between scientist and politician, but I had honestly believed that a modern understanding had come to pass acknowledging the necessary degree of autonomy and impartiality relegated to scientific and regulatory bodies (like Parks Canada).  It is this fundamental perversion of the relationship between policy and empirical study which, to me, is the most disturbing aspect of Harper's leadership.


* For anyone who voted for the Conservatives because they were sick of a minority government leading to constant elections, that's a terrible argument.  Sometimes having a non-functional Parliament is better than having one which will toe the line for bad policy.

Saturday, May 12, 2012

Solution to Puzzle 16: Trip to Canada

Here is the solution to the "Trip to Canada" puzzle:

The Uncle traveled to one location in each province and territory in Canada in the following order:

1.) Alert, Nunavut
2.) Yellowknife, Northwest Territory
3.) Whitehorse, Yukon
4.) Whistler, British Columbia
5.) Red Deer, Alberta
6.) Moose Jaw, Saskatchewan
7.) Winnipeg, Manitoba
8.) Newmarket, Ontario
9.) Trois-Rivieres, Quebec
10.) Eel River Crossing, New Brunswick
11.) Cow Bay, Nova Scotia
12.) Cornwall, Prince Edward Island
13.) Labrador City, Newfoundland and Labrador

I unfortunately did not realize that there is also a Cornwall, Ontario, until after I published this puzzle, but I hope the rest of the locations were unique enough for the puzzle to still have been solvable (plus, Cornwall, PE is the only one which would make sense given the order of the Uncle's travels).