5 ms·

Deep belief image recognition on Raspberry Pi

80 points by liquidmetal 11y ago

bigiain 11y ago
Apart from a recent (26 day old) update to the javascript, this looks like it got abandoned back when Google acquired Jetpac (who's website now seems defunct). Anyone know what the status of this is? I guess there must've been some more-recent-than_aug-2014 updates, since it mentions RaspberryPi2s...
- choppaface 11y ago
  Believe Pete Warden ( http://petewarden.com/ http://petewarden.com/ ) works on TensorFlow now.
- nl 11y ago
  http://petewarden.com/2015/05/10/image-recognition-on-the-raspberry-pi-2/ http://petewarden.com/2015/05/10/image-recognition-on-the-ra... http://petewarden.com/2015/05/18/jetpacs-deep-learning-framework-on-the-beaglebone-black/ http://petewarden.com/2015/05/18/jetpacs-deep-learning-frame... https://github.com/tensorflow/tensorflow/issues/445#event-486230566 https://github.com/tensorflow/tensorflow/issues/445#event-48...
dharma1 11y ago
you could also try this: https://github.com/teradeep/demo-apps/tree/master/generic-embedded https://github.com/teradeep/demo-apps/tree/master/generic-em... or this: http://mxnet.readthedocs.org/en/latest/tutorial/smart_device.html http://mxnet.readthedocs.org/en/latest/tutorial/smart_device... Not really RPi related - but I think it's only a matter of time until we have mobile GPU optimised BLAS libraries working with Mali, Adreno etc. At that point machine learning will become more feasible on current mobile devices
buserror 11y ago
Had had a good play with that lib earlier on debian, it's pretty cool! Very lightweight (considering) and I tossed it a few pictures of stuff, and it's quite impressive. I particularly liked my eBay shopping history. Didn't like my landscape photos so much, but it's likely to be that it hasn't been trained on any... What would be fantastic is a way to recover some sort of rough coordinates of what the classification have found.
- dharma1 11y ago
  bounding box - https://github.com/rbgirshick/py-faster-rcnn https://github.com/rbgirshick/py-faster-rcnn
es0m 11y ago
This lib is so small and self-contained that it was straightforward to port it to Windows (Visual Studio): https://github.com/jetpacapp/DeepBeliefSDK/pull/62 https://github.com/jetpacapp/DeepBeliefSDK/pull/62 I had trouble getting Caffe and Torch to run natively (i.e. MS toolchain), anyone knows of a good Windows port for these that one can link to?

80 points by liquidmetal 11y ago

Apart from a recent (26 day old) update to the javascript, this looks like it got abandoned back when Google acquired Jetpac (who's website now seems defunct). Anyone know what the status of this is? I guess there must've been some more-recent-than_aug-2014 updates, since it mentions RaspberryPi2s...

choppaface 11y ago
Believe Pete Warden ( http://petewarden.com/ http://petewarden.com/ ) works on TensorFlow now.
nl 11y ago
http://petewarden.com/2015/05/10/image-recognition-on-the-raspberry-pi-2/ http://petewarden.com/2015/05/10/image-recognition-on-the-ra... http://petewarden.com/2015/05/18/jetpacs-deep-learning-framework-on-the-beaglebone-black/ http://petewarden.com/2015/05/18/jetpacs-deep-learning-frame... https://github.com/tensorflow/tensorflow/issues/445#event-486230566 https://github.com/tensorflow/tensorflow/issues/445#event-48...

dharma1 11y ago

you could also try this: https://github.com/teradeep/demo-apps/tree/master/generic-embedded https://github.com/teradeep/demo-apps/tree/master/generic-em... or this: http://mxnet.readthedocs.org/en/latest/tutorial/smart_device.html http://mxnet.readthedocs.org/en/latest/tutorial/smart_device... Not really RPi related - but I think it's only a matter of time until we have mobile GPU optimised BLAS libraries working with Mali, Adreno etc. At that point machine learning will become more feasible on current mobile devices

buserror 11y ago

Had had a good play with that lib earlier on debian, it's pretty cool! Very lightweight (considering) and I tossed it a few pictures of stuff, and it's quite impressive. I particularly liked my eBay shopping history. Didn't like my landscape photos so much, but it's likely to be that it hasn't been trained on any... What would be fantastic is a way to recover some sort of rough coordinates of what the classification have found.

dharma1 11y ago
bounding box - https://github.com/rbgirshick/py-faster-rcnn https://github.com/rbgirshick/py-faster-rcnn

es0m 11y ago

This lib is so small and self-contained that it was straightforward to port it to Windows (Visual Studio): https://github.com/jetpacapp/DeepBeliefSDK/pull/62 https://github.com/jetpacapp/DeepBeliefSDK/pull/62 I had trouble getting Caffe and Torch to run natively (i.e. MS toolchain), anyone knows of a good Windows port for these that one can link to?