It seems that most of us have missed the importance of this robot. I was fortunate enough to see the research group present for this and another of their robots (which does the same thing but is large enough to carry a substantial cargo as well). This robot autonomously digitizes large enviorments including texture maps on it. While admittidly there is room for improvement it is more than just an important step. It could be snuck up into all sorts of places people can't fit and be used to search through rubble, or be used to search through spaces in the pyramids too small for people to fit through. Similar technology could be used in space probes sent to places like mars to digitize the enviorment where human control over them takes an extremely long time, or on the bottom of the ocean giving us an image of it we didnt otherwise have. While similar robots have existed this one pans the time of flight laser scanner in order to digitize the entire room, and while it doesnt do it super quickly I saw a video of it working and it does work fairly fast. It should be able to digitize a decent size room in 3 or 4 minutes fairly completely.
I heard their presentation on it at the Recent 3DIM conference. It uses a time of flight laser scanner, but their addition to it is that it's on a tilt angle which allows them to take a wider range of images. The Time of flight laser they use is acurate to about 2 CM
Sorry I wasnt specific in what I was saying. I was trying to dumb it down a little. My claim of 500 was meant towards the number of individuals involved in the studies, not the number of images. I believe the number of individuals is MUCH more important in determining the effectiveness in a real world border crossing type situation than the number of images involved.
While your correct the feret database had more than that (1200 individuals in 2000 I even double checked it witht the feret website) they're acuracy was around 95% far lower than most of the modern claims today (Yes there has been considerable advancements) with a false alarm rate so high there would be tens of thousands of false alarms in airports like Ohare or LAX. So, I didnt include the feret study in what I mentioned. Sorry I was trying to dummy it down. It was my fault.
Personally, I think the problem with 2D face recognition is simply that there isnt enough data in the standard 2D image to differentiate amongst millions of people. Further, in all of the FERET studies or any of the studies involving large groups these are willing subjects. The subjects arent attempting to disguise their identies through beards, or glasses. Two things which will kill completely most modern techniques. While some smaller studies have been done on this issue they have shown not very promising results at least with current software and methods. Therefore, I still believe that despite some promising stuff going on right now we're quite far away from the security camera having any chance of being to identify who you are.
As someone who has been doing research in areas of computer vision, and specifically identification and a member of a Computer Vision Research Laboratory, I just thought I would make a few comments here.
Some area's of computer vision, in relation to big brother, have been around for a while and actually work quite well already. These areas include but are not limited to fingerprint, iris, and hand just to name a few. Those mentioned above are already in commercial applications around the country used for everything from secure entry into the country at immigration stations, to secure entry into rooms/labs/whatever, and to confirm identification for logins to computer and other systems. They work well (always some room for improvement), but require a completely willing subject and carry a certain 'stigma' of big brother and criminals with them that makes them less viable.
The view mentioned here that researchers want to work towards is having a standard camera (like a security camera) able to identify people. However, despite some claims so far (most recent interesting claim out of Isreal), so far no one has proved to have ANYTHING that would be viable in a real world application. Best systems thus far have never even been tested with a database of over 500 people, most significantly less than that, and tend to not work well over time. Usually, they work fairly well the same day and then exponentially decrease in their effectiveness until around 6 months when you may as well be randomly guessing because you'd do about as well as most algorithms. Overall, I don't think you have anything to fear from big brother here anytime soon.
Slashdot Mirror
It seems that most of us have missed the importance of this robot. I was fortunate enough to see the research group present for this and another of their robots (which does the same thing but is large enough to carry a substantial cargo as well). This robot autonomously digitizes large enviorments including texture maps on it. While admittidly there is room for improvement it is more than just an important step. It could be snuck up into all sorts of places people can't fit and be used to search through rubble, or be used to search through spaces in the pyramids too small for people to fit through. Similar technology could be used in space probes sent to places like mars to digitize the enviorment where human control over them takes an extremely long time, or on the bottom of the ocean giving us an image of it we didnt otherwise have. While similar robots have existed this one pans the time of flight laser scanner in order to digitize the entire room, and while it doesnt do it super quickly I saw a video of it working and it does work fairly fast. It should be able to digitize a decent size room in 3 or 4 minutes fairly completely.
I heard their presentation on it at the Recent 3DIM conference. It uses a time of flight laser scanner, but their addition to it is that it's on a tilt angle which allows them to take a wider range of images. The Time of flight laser they use is acurate to about 2 CM
While your correct the feret database had more than that (1200 individuals in 2000 I even double checked it witht the feret website) they're acuracy was around 95% far lower than most of the modern claims today (Yes there has been considerable advancements) with a false alarm rate so high there would be tens of thousands of false alarms in airports like Ohare or LAX. So, I didnt include the feret study in what I mentioned. Sorry I was trying to dummy it down. It was my fault.
Personally, I think the problem with 2D face recognition is simply that there isnt enough data in the standard 2D image to differentiate amongst millions of people. Further, in all of the FERET studies or any of the studies involving large groups these are willing subjects. The subjects arent attempting to disguise their identies through beards, or glasses. Two things which will kill completely most modern techniques. While some smaller studies have been done on this issue they have shown not very promising results at least with current software and methods. Therefore, I still believe that despite some promising stuff going on right now we're quite far away from the security camera having any chance of being to identify who you are.
As someone who has been doing research in areas of computer vision, and specifically identification and a member of a Computer Vision Research Laboratory, I just thought I would make a few comments here. Some area's of computer vision, in relation to big brother, have been around for a while and actually work quite well already. These areas include but are not limited to fingerprint, iris, and hand just to name a few. Those mentioned above are already in commercial applications around the country used for everything from secure entry into the country at immigration stations, to secure entry into rooms/labs/whatever, and to confirm identification for logins to computer and other systems. They work well (always some room for improvement), but require a completely willing subject and carry a certain 'stigma' of big brother and criminals with them that makes them less viable. The view mentioned here that researchers want to work towards is having a standard camera (like a security camera) able to identify people. However, despite some claims so far (most recent interesting claim out of Isreal), so far no one has proved to have ANYTHING that would be viable in a real world application. Best systems thus far have never even been tested with a database of over 500 people, most significantly less than that, and tend to not work well over time. Usually, they work fairly well the same day and then exponentially decrease in their effectiveness until around 6 months when you may as well be randomly guessing because you'd do about as well as most algorithms. Overall, I don't think you have anything to fear from big brother here anytime soon.