11–20 of 33 posts
tried two images: http://kephra.de/Dampf/IMG_20140620_133839_800x600.jpg http://kephra.de/pix/Snoopy/thump/IMG_20130822_135928_640x48... Tried some other images from the lifestyle section of my homepage, but it looks as if the system newer saw a sewing machine before as it gives "Low recognition confidence", and no tags.
Earlier quoted context omitted.
AFAIK, it's using a Deep Neural Network; which means, the inputs are, basically, pixel values (possibly normalized), and all feature detection, etc. is done in the layers of the network.
yep, they try to learn an image's high level features by learning an autoencoder (that is a transform that takes an image and tries to produce the same image) via a sandglass shape multi layer network. Here is a very readable paper by Hinton himself that describes the approach: http://www.cs.toronto.edu/~hinton/science.pdf
Earlier quoted context omitted.
AFAIK, it's using a Deep Neural Network; which means, the inputs are, basically, pixel values (possibly normalized), and all feature detection, etc. is done in the layers of the network.
yep, they try to learn an image's high level features by learning an autoencoder (that is a transform that takes an image and tries to produce the same image) via a sandglass shape multi layer network. Here is a very readable paper by Hinton himself that describes the approach: http://www.cs.toronto.edu/~hinton/science.pdf
This is the implementation: http://torontodeeplearning.github.io/convnet/
Didn't give me results at all to the three images I uploaded. Might be broken.
Earlier quoted context omitted.
yep, they try to learn an image's high level features by learning an autoencoder (that is a transform that takes an image and tries to produce the same image) via a sandglass shape multi layer network. Here is a very readable paper by Hinton himself that describes the approach: http://www.cs.toronto.edu/~hinton/science.pdf
Could it maybe be worthwhile to augment the data with simple image features? E.g. the human visual system is believed to rely on high-level/top down as well as on local/bottom up features (although that might also be simply because of the necessity to compress things for the low nerve count in the optical nerve).
That being said: even deep learning requires some sort of feature engineering at times (even if its pretty good with either hessian free training or pretraining).
The main thing with images is ensuring scaling them.
The trick with deep belief networks in particular is to make sure the RBMs have the right visible and hidden units (Hinton recommends Gaussian Visible, Rectified Linear Hidden).
Happy to answer other questions as well!
LOL! http://i.imgur.com/Xs3GrGk.png
Are there actually any image feature detectors and descriptors involved (like blob, edge and texture detectors) or is this solely based on artificial neural networks?
Are there actually any image feature detectors and descriptors involved (like blob, edge and texture detectors) or is this solely based on artificial neural networks?
Interestingly it has been shown that the result from some neural networks is equivalent to using classification with some predefined filters. These filters could be considered as a feature descriptor. See this talk from CVPR http://techtalks.tv/talks/plenary-talk-are-deep-networks-a-s... .
Also can it tell you where in the image the identified object is?