Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : vsnx -- Ignore unavailable to you. Want to Upgrade?

To: ChinuSFO who wrote (14)	12/11/2001 10:24:26 AM
From: Green Receipt	Respond to of 41

It's better than nothing. And guess what, still no response from VSNX.... It's not good to ignore your customer... Their algorithm returns a score ranging from 0.0 to 10.0. Often we have found their engine returns the wrong hit (person). Moreover, we had a picture we took and used it against one of their demo test software programs and it failed to 'match the person correctly. I then sent the picture to them where they could test with their system and I was told over the phone, that their algorithm, with their direct software id'd the wrong person. They will tell you that the lighting conditions make all the difference in getting good quality images of people. However in the general open public cases, lighting will be subpar at best and that will seriously affect the ability to make correct judgements. What it boils down to is you still need people in the loop. If the system someone develops leaves people out, then u can pretty much guarantee the wrong people will be identified. There is no worse advertising that can be had than to use their 'screensaver demo' and have it think a total stranger is the authorized person who has authority to use the computer. That scenario actually happened to us in June when we were demoing it. In their screen saver demo, they take 8 or more images of the authorized person. and then each time in lock mode a face is found, it's compared against the 8 saved images. about the raytheon deal. I interviewed with a biometrics group at raytheon earlier this year and even though I was asked to work there I turned them down. At the time they didn't seem to have much of an idea where they were heading or what they were doing. Who knows if this is the same group or not, and perhaps they now have a better idea. But based on my experience, I'll be happily suprised if the raytheon/visionics system works as promised. Before my current biometrics work I was involved in writing software (as an outside contractor) that was supposed to run on the T3 (http://www.raytheon.com/press/1996/sep/socom.html) I had much trouble and finally with the help of some other engineers were able to figure out their T3 VME bus was at fault. We were never able to get anyone at Raytheon to own up to the problem and around the time I quit working on that effort (connecting ITOS (http://www.disa.mil/pubs/circulars/dc800701basic.html ) to run on a T3) the army had basically yanked the contract from Raytheon (or something like that). I've disgressed and strayed off topic... Back to the topic. If Raytheon and Visionics pull this off then more power to them, however based on what I have seen first hand, I'm doubtful they will ever produce a working sellable system. I'd love for them to prove me wrong. And I have friends that work at Raytheon in different areas. I'm sure this isn't a problem with all of raytheon, but just some of the groups.