Thursday, November 27, 2008

Showing Girls Careers They Could Love

While catching up on Slashdot, I found this question asked:
"My niece just took the ACT and got a perfect score on the math section. 25 years ago, when I took the test, the kids who aced the math section were pretty special. Her score, combined with straight A's so far in high school, suggest to me that she might be able to go to a top university (MIT?) based on her math aptitude. The rub is that she doesn't like math or science, even though she finds them easy. She doesn't want to be an engineer or scientist. I thought the folks here would be a great group to ask: What are some creative, not too nerdy professions that nonetheless require a talent for math, engineering, or science?"
I was immediately reminded me of some of the discussions held at the National Conference on Women in Engineering last week. For example, one of the key themes touched upon was the fact that girls may not become interested in science and engineering because they see it as too focused on the tools themselves (computers, wrenches, electronics, whatever) rather than what can be done with them. This goes a bit further than the usual nerdy image and similar kinds of turn-offs.

The comments for the Slashdot article spent some time emphasizing that someone with good math or science skills shouldn't be pushed into doing a career they don't want to do just because of these abilities.

A few examples:
I'm from the UK and I suffered the same fate that you wish to throw upon your daughter - being coerced into a specific degree program at a top London university just because I excelled in that area in my secondary (high) school, without realising myself what a change it would be from the material I had learnt thus far through my life. It certainly didn't do me any favours.

I myself was pressed into (natural) 'science' because math was easy to me, which in the long run (decades) turned out to be a major desaster that I am still trying to recover from.

I agree - my sister was nearly pressured into an engineering route at college by schooling and sponsorship deals but stuck to her guns and has a postgraduate diploma in music performance on two instruments. She's very happy - she can do the music when the work is available for her instruments, and to fill in of the time can get "technical" positions in sales/marketing for engineering companies.

You are better off asking her what she wants to do. What is she interested in? If she has no idea then going to a large university where she'll be exposed to a number of different fields and opportunities is not a bad idea.

In fact, I do not disagree with this sentiment - nobody should be pushed into a certain field because others perceive it as being right for them.

However.

It is worth our time to help young women see what these fields are really all about. If those who might have an aptitude for science or engineering never get a chance to see the real story - outside of the nerdiness, or the focus on tools - they may never discover that they also have a real love for it.

I think this is largely what the original question might really be about: How do I show my niece the amazing things scientists and engineers do to help society? What do they do with all these tools that the fields are so well known for? What makes these careers so fulfilling? By giving the niece the power of this information, perhaps she will make a more informed decision. And if that decision doesn't happen to be science or math or engineering, then that's ok.

Wednesday, November 26, 2008

Scratch That

Yet another cool interface comes out of Carnegie Mellon: scratch input. Using nothing more than a microphone whose input is filtered to remove lower frequency sounds (like voice), scratches on surfaces like walls or tables can be captured and analyzed as gestures.


I am amazed at how well this works on regular walls in the home. The video demonstrates that scratches made in the corner and above/beside doors work just as well as those made right beside the mic!

The demo shows a user controlling a music player with his scratch gestures, which is definitely cool, but I'm wondering if this might be better employed in the attempt to recreate those Minority Report immersive environments. Why use video to capture human hand movements as gestures, which could arguably require much more resolution, when some simple sounds would suffice? It does mean the user would be required to actually touch a surface to interact with it, but let's face it, that's much more natural than waving our hands in the air anyway.

Sunday, November 23, 2008

NCWIE: Canadian WISE Groups

Barb and I presented our talk yesterday to around 40 delegates of the National Conference on Women in Engineering, held in London, Ontario. I'm extremely happy with how well it turned out. We started with a video slide show I made (complete with dramatic music to open, followed by dance music to keep the crowd's interest). We made sure to bring the audience into the discussion many times, and kept their interest by throwing in more fun videos to soften the dullness of the more technical details. We received amazing feedback from multiple people!

But the one thing I'm the happiest about is a new wiki I set up thanks to the suggestions of our audience. We had so much great discussion that could have lasted an entire day on its own. It would be a shame to cut that off without finding a way to continue sharing ideas.

The wiki is called Canadian WISE Groups and, I'm hoping, will be a space that WISE groups from all across our nation can list their contact info and share ideas on how they run their clubs.

There have been similar initiatives that I have heard about recently, but none so far have seemed to include both scientists and engineers, or they haven't covered the whole country.

If you know about a group that supports women in science and/or engineering, please consider requesting a membership to the wiki and adding it. There is also a list of links to organizations that support women, and it's not restricted to Canadian initiatives. If you find that your organization is not there, definitely feel free to join the wiki and add it! This site can only grow and be useful if you help build it.

Saturday, November 22, 2008

NCWIE: Entrepreneurship

Fellow CU-WISE Executive member Barbora and I are currently in London, Ontario for the National Conference on Women in Engineering, or NCWIE. We've been to conferences like this before (like the Grace Hopper Celebration of Women in Computing), but this time we have come as speakers!

I made it to a couple of talks yesterday, but the one I remember the most was about entrepreneurship. Jennifer MacDonald appeared on CBC's version of the Dragon's Den to pitch her organic salad dressing that contained flax seed oil and didn't taste terrible because of it. She did fairly well in the Den, got a deal, and is now looking at options for selling her company after growing it with her investors (and she'll continue to profit as a result, of course).

She spoke with such enthusiasm that my small thoughts of one day having my own business turned into a real desire. This is partly thanks to Barb, who very much want to work with our CU-WISE executive team to make something happen (since we already work so well together). I have no idea what kind of business I'd like to do (design a product? provide a service? consult?), but I will now be watching for ideas and some kind of need in the market. Perhaps I should start reading books about the topic, such as ZAG, a title about branding recommended by Jennifer.

If that great idea ever does strike, I'm happy to know that there is help to get started. Take Women 2.0:
Women 2.0 is committed to increasing the number of women entrepreneurs starting high growth ventures by providing the resources, network, and knowledge for the launch and growth of their company.

Our vision is to be a catalyst for change, mobilizing a global community of ambitious women entrepreneurs seeking to advance the world through technology.
Maybe we'll even be able to enter one of their startup competitions:
Despite the downturn in the economy, many entrepreneurs and VCs are saying that now is a great time to start a company. The Women 2.0 Pitch 2009 competition can get you started. Here is your chance to PITCH.
In the meantime, I'm glad to know that my potential career path of becoming an instructor at university might support the idea of developing a start-up during the summer months off. The future feels very exciting...

Friday, November 14, 2008

CU-WISE in CarletonNOW

The November 2008 issue of CarletonNOW includes an article about Carleton's Women in Science and Engineering! I've copied it below for your convenience, but the original is also available online.

Teaching > Carleton students CU-WISE
Posted Nov. 10/08
By Heather Montgomery

At the beginning of her graduate studies in 2007, Barbora Dej was looking for a way to connect to other women at Carleton who shared her interests. As a woman, Dej is a minority in her chosen field of engineering.

"I wanted to know, is there something at Carleton that supports women in engineering," says Dej. She scoured the Internet and eventually found the Carleton University branch of Women in Science and Engineering (WISE).

Last year CU-WISE, a student branch of the Ottawa WISE chapter, was revived by a dedicated team of women. The group brings together women from across all disciplines of science and engineering, for networking, discussions and social change.

Left to right: Barbora Dej, Natalia Villanueva-Rosales, Serena Ngai and Gail Carmichael
participate in Carleton’s branch of Women in Science and Engineering known as CU-WISE.


"We’re not necessarily trying to increase the numbers," explains Gail Carmichael, a master’s student in computer science who is also on the CU-WISE executive for internal affairs.

"We’re trying to make sure that people who would be interested originally wouldn’t see those barriers and would try it out and not be afraid to do so."

The executive of CU-WISE focuses on staying organized. They exchange anywhere from 20 to 100 e-mails a week. They maintain their website and blog meticulously. For them, being able to have that virtual connection with potential members and interested people is a priority.

"Everything’s on the website," says Natalia Villanueva-Rosales, a PhD candidate in computer science who’s on the executive of the CU-WISE virtual communications committee.

"Every single detail, we take care of it."

Before its revival last year, CU-WISE was a struggling club at Carleton. Now, the group has over 100 members and is involved in hosting and participating in various events throughout the year. For a lot of the members, just knowing that there were many women like them out there was reassuring.

In September, CU-WISE hosted an event for first-year women that provided them with tips about how to survive their undergraduate degrees, and what to expect as a woman in science or engineering.

"Seeing all these first-years is sort of like seeing yourself," says Carmichael. "We all wish we’d had CU-WISE."

The women of CU-WISE also believe in social connection. They recently travelled to the Grace Hopper conference in Colorado to meet with women from all over the world.

"That’s one big reason women don’t get into computer science and engineering," says Carmichael. "They don’t understand that there is that social impact you can have as well. They imagine being in a cubicle all day coding. But there’s so much you can do."

Whether it’s one girl or 100, the women of CU-WISE want to let all women at Carleton know they’re not alone. That’s what drives them every day.

"If you make the difference in the life of a girl who doubted whether she’d come into something like computer science or engineering or if she was thinking of leaving the school and she stayed, that’s worth it," says Villanueva-Rosales.

On Nov. 26, CU-WISE is hosting a lecture by Carleton president and vice-chancellor, Dr. Roseann O’Reilly Runte.

Wednesday, November 12, 2008

Hands-Free Computing

Everyone loves to talk about the future of computing and its user interfaces. For many people, this future includes Minority Report multi-touch interfaces. Beautiful and seemingly functional, these immersive environments are certainly attractive. How close are we to achieving something similar with current technology?

Johnny Chung Lee brought us affordable multi-touch using the popular Wii remotes. On the more pricey end of the spectrum, Microsoft brought us multi-touch tabletop computing. One of the iPhone's greatest features (along with related products) is its intuitive multi-touch interface. All of these examples are great, but none seem to come close to slick interface used by Tom Cruise's Minority Report character.

Perhaps Mgestyk, the newest player in the game, has managed to bring us a little closer?

According to Gizmodo:
We've seen gesture controls in gadgets before, but Mgestyk Technologies wants to bring them to your home PC. Using only a 3D camera and proprietary software, the Mgestyk gesture control system is able to capture small hand movements and translate them into commands. These commands can be applied to almost any windows application, including video games. Judging from the clips they have on their site, the system seems to work as advertised, though there does appear to be a little lag. Pricing is expected to be within the range of a high end webcam which by our estimates is around $150.
You can see the gestures in action for various games and apps in the below video.



To me, this isn't really anything too cutting-edge, but I did enjoy seeing how gesture-based gaming might look. I've concluded that, while it seems really cool to be able to aim and shoot with your own hands, I can't see anyone wanting to do it for long. If holding your shoulders tight to use your mouse can cause such havoc on your body, can you imagine having to hold your hands out in free space to play a game for a couple of hours?

I think the real potential for gesture-based computing lies in the area of "step-up and use" computing, like tourist kiosks. These systems generally aren't used for very long and should be very quick to learn. A well crafted interface would make use of a small number of intuitive gestures that even your grandma could understand. Time will tell whether anyone will be able to pull this off.

Monday, November 10, 2008

Community Photo Collections

If you've never taken the time to check out Photosynth, do it now. Click your way to the Sphinx example and watch what happens when you drag your mouse around the main window. You will see a set of points that look an awful lot like some pyramids and the Sphinx. You can zoom in and start seeing some photographs overlaid onto the point cloud. What's so impressive about this, you ask? Well, all this data was obtained from photographs alone, reconstructed into this 3D navigable wonder you see before you. Pretty cool.

There are times when you might have found yourself a little shutter-happy when visiting some exotic, far-off land. Yet you probably still don't have enough photos of the exact same thing to recreate something like the Sphinx model. Fortunately, photo-sharing sites like Flickr help solve that problem! Starting with tags and geo-coding, and double checking by matching the photos to each other, we can find all the photos we need. That's exactly what researchers at Washington University did for the Community Photo Collections project, which became the base of Photosynth!

From the site:
With the recent rise in popularity of Internet photo sharing sites like Flickr and Google Images, community photo collections (CPCs) have emerged as a powerful new type of image dataset for computer vision and computer graphics research. With billions of such photos now online, these collections should enable huge opportunities in 3D reconstruction, visualization, image-based rendering, recognition, and other research areas. The challenge is that these collections have extreme variability, having been taken by numerous photographers from myriad viewpoints with varying lighting and appearance, and often with significant occlusions and clutter. Our research seeks to develop robust algorithms that operate successfully on such image sets to solve problems in computer vision and computer graphics.

For those with a few extra minutes (sixty-one extra minutes, to be exact), you can get a pretty good sense of the technology behind the project in this Google Tech Talk. The presentation isn't too technical, and uses some great demos to show what's going on. For the geeks out there, there is structure-from-motion source code available, too!



A lot of what's going on behind the scenes is related to what I'm trying to do with my thesis. We both need to find matches between two photos (or a panorama and a photo in my case), and recover the geometry between these entities. For applications like Photosynth, this geometry is used to obtain a 3D reconstruction of the scene (that's what all those points were in the Sphinx example). I might want to use the geometry to add virtual objects, like historical buildings or geographical information, to the photograph, using the known configuration of the panorama.

Thursday, November 6, 2008

Thesis Update: Geometry Between Cubes and Photos

This post is for anyone wondering how my thesis has been going. It's a bit more technical than some of my other stuff. I am presenting some of the more important results with explanations of the methodology on my personal website.

First off, I'd like to share a description of my research that I used when applying for a PhD scholarship to ensure we're all on the same page (because the exact research goals and purpose change often in my mind):
Spherical panoramas have been used in such high-profile applications as Google's Street View to allow users to naturally explore real-world images from the comfort of their own homes. In the case of Google, crude street information is augmented onto panoramic images to aid navigation both while viewing the panorama
and, in theory, while driving or walking on location. It may be useful to augment a photograph (taken with a cell phone for example) of an intersection or tourist attraction while a person is actually standing there, but an exact camera location would be required to do this.

One way to obtain this camera location is to compare the photograph with nearby spherical panoramas (which can be found using a rough GPS location estimate). If the panoramas have been captured and saved with positional information, then the scene geometry between the user's camera and the panoramas will help recover information about that camera's position, thereby allowing for an accurate augmentation.

Previous work [1] has established a method to recover position information between two panoramas, and the theory established there may be applicable to this case of comparing a photograph with a panorama. This will be verified during the course of the research, but the thesis will mainly investigate the best way to efficiently
obtain a large number of match correspondences between the photograph and the panorama, as this is the first step to finding the mathematical structures that describe the geometry. The format of the panoramas in [1] is that of a cube. Because of the 'seams' along adjacent faces, some feature points may not find correspondences to the same features visible in the planar photograph. As such, this
format will be compared with a cylindrical representation of the panoramas, which has no seams but must deal with curvature issues, to see if more correspondences might be found.

In addition, it must be determined what information should be stored on a central server along with the set of pre-captured panoramas. As much work as possible should be pre-computed to ensure the user's photograph is sent back with an augmentation as soon as possible.

While road information such as that augmented onto images in Street View may not need to be highly accurate, there are many other applications that would require more precision. For example, virtual objects or textual information could be added to the photograph before it is sent back to a tourist learning about a historically significant area. In a case like this, a natural augmentation obtained with an accurate camera location is all but essential.

---
[1] Kangni, F. and Laganiere, R. (2007) Orientation and Pose recovery from Spherical Panoramas. ICCV

Basically, I'm trying to figure out the best way to find matches and/or an essential matrix between a cubic panorama and a photograph.

So far, I have taken the theory from [1], which explained how to find an essential matrix between two cubic panoramas, and modified it to work with a cubic panorama and a photograph. The trick here was that the photograph would not be calibrated (i.e. we don't know the camera's properties like its focal length). Usually, this would mean that we'd want to find a fundamental matrix instead. However, this would force us to abandon the advantageous ability to consider all faces of the cubic panorama at the same time (by using the normalized 3D coordinates of points on the face images). We would have to match each face individually, and once we found the face with the most matches, find a fundamental matrix between it and the photo. It would seem that using points on more than one face would help us get more matches and a more accurate result.

Instead, I looked into the possibility of using calibrated points from the cubic panorama and uncalibrated image points. The resulting matrix to find would be a cross between the calibrated essential matrix and the uncalibrated fundamental matrix. The basic idea is informally presented here, and I call it a "pseudo-essential matrix."

By hand picking some matches between some panoramas and photos, I was able to ensure that the pseudo-essential matrix idea was sound. Some initial results showing this are available here. The only major issue seen here was the instability around the epipoles.

I am currently working on improving the ability to find a pseudo-essential matrix automatically. The early progress can be seen here. Many of the matches found after the nearest-neighbour thresholding appear to be correct, but a pseudo-essential matrix is not found. I need to check whether the quality measures are too strict, and perhaps evaluate how my RANSAC algorithm is working.

By the time I am finished my research, I don't think that the concept will be ready for using on consumer mobile devices. But it would be really cool to see it used as a starting point for the next great mobile app! With the resources available at places like Google, I'm convinced it's doable.