I built a DIY license plate reader with a Raspberry Pi and machine learning
towardsdatascience.com
I built a DIY license plate reader with a Raspberry Pi and machine learning
1–10 of 78 posts
Re: I built a DIY license plate reader with a Raspberry Pi and machine learning
#2The video referenced in the blog:
https://www.youtube.com/watch?v=gsYEZtecXlA
mentions that they used 20 K80 instances to get real time inferencing. That seems a bit excessive IMO for an object detector.
Re: I built a DIY license plate reader with a Raspberry Pi and machine learning
#3The solution described in the blog uses 12 Nvidia T4s, which are basically specialized RTX 2080s. Depending on what vCPUs etc. get used, it's running 1000-1500 W of computing power continuously.
If their car was electric, this project would be increasing the power consumption of the vehicle by around 10%.
Re: I built a DIY license plate reader with a Raspberry Pi and machine learning
#4The solution described in the blog uses 12 Nvidia T4s, which are basically specialized RTX 2080s. Depending on what vCPUs etc. get used, it's running 1000-1500 W of computing power continuously. If their car was electric, this project would be increasing the power consumption of the vehicle by around 10%.
I may have missed something, but I believe the T4s are running in the cloud, so it wouldn't have any effect on the car's power consumption.
Re: I built a DIY license plate reader with a Raspberry Pi and machine learning
#5It’s just a question of time before people can hack their own similar facial recognition system. The new glasses from Bosch, with a projector straight on to the retina, is an obvious choice for displaying personal info about everybody in reach of the camera. With 5G, everybody has a reliable access to all the computing power needed.
Re: I built a DIY license plate reader with a Raspberry Pi and machine learning
#6The solution described in the blog uses 12 Nvidia T4s, which are basically specialized RTX 2080s. Depending on what vCPUs etc. get used, it's running 1000-1500 W of computing power continuously. If their car was electric, this project would be increasing the power consumption of the vehicle by around 10%.
I may have missed something, but I believe the T4s are running in the cloud, so it wouldn't have any effect on the car's power consumption.
To clarify, I mean indirectly the power consumption. I guess this will become an important point of discussion soon-ish, as we get more and more electric cars and ML stuff in them.
My view is, if your car uses electrons, and some of those are for compute, and you just offload the compute to the cloud, you haven't actually reduced the total electricity consumption.
Similarly here, the total "footprint" of the car is increased by adding this feature.
Re: I built a DIY license plate reader with a Raspberry Pi and machine learning
#7It’s just a question of time before people can hack their own similar facial recognition system. The new glasses from Bosch, with a projector straight on to the retina, is an obvious choice for displaying personal info about everybody in reach of the camera. With 5G, everybody has a reliable access to all the computing power needed.
5G is not required for that.
Re: I built a DIY license plate reader with a Raspberry Pi and machine learning
#8Earlier quoted context omitted.
I may have missed something, but I believe the T4s are running in the cloud, so it wouldn't have any effect on the car's power consumption.
To clarify, I mean indirectly the power consumption. I guess this will become an important point of discussion soon-ish, as we get more and more electric cars and ML stuff in them. My view is, if your car uses electrons, and some of those are for compute, and you just offload the compute to the cloud, you haven't actually reduced the total electricity consumption. Similarly here, the total "footprint" of the car is i…
To be honest, there are open source ALPR (https://github.com/openalpr/openalpr) that could probably run lower framerates even on modern i7 processors (http://doc.openalpr.com/getting_started.html#hardware-requir..., https://docs.google.com/spreadsheets/d/1FNwEuJAgZ1LyM2GGd7VR...)
Re: I built a DIY license plate reader with a Raspberry Pi and machine learning
#9Plenty of room for optimization. Raspberry PI has enough power to detect license plate numbers on its own, offline.
Re: I built a DIY license plate reader with a Raspberry Pi and machine learning
#10The video referenced in the blog: https://www.youtube.com/watch?v=gsYEZtecXlA mentions that they used 20 K80 instances to get real time inferencing. That seems a bit excessive IMO for an object detector.
I'm the creator here. Yeah, 20 K80s is a bit excessive. That's because cortex (cortexlabs), which is the ML-model-deployment platform didn't initially have multiprocessing on each of their replicas - so I was bound to using just one CPU per GPU. AWS has instances with 4, 8, 16 vCPUs and so on.
Once cortex started supporting gunicorn (still yet unreleased but present on their master branch), I was able to reduce the number of GPUs significantly. Finally, the detector only needs 2 T4 GPUs and the identification part is the most expensive one (10 T4s GPUs) for a grand total of 12.
Converting the models to use single-precision could further reduce the need to about 1.5 GPUs for T4s or just 1.2 GPUs for V100 - which in both cases it would still mean using 2 GPUs.