0

I'm using the default 'bvlc_reference_caffenet' model. I'm trying to detect a spatula. Now the results I'm getting are pretty satisfactory. The spatula class is always among the top 5 predicted classes but the rest are useless random things that I'm never going to be looking for. I could add a filter at the end to remove undesirable results but does Caffe provide this functionality on it's own? Can not look for said classes?

legolasshegolas
  • 1
  • 2

1 Answers1

0

Yes, it does. 'bvlc_reference_caffenet' comes with a text file that defines structure of the neural network. It is composed of inuput layer, set of hidden layers and output layer. if you'd like to make it the best possible spatula-finder, then you have to modify the output layer and make it "spatul" and "rest of the world".

Mind that it requires re-training the model. In fact it's enough if you just refine the model by taking weights of the existing model and going only through fraction of iterations that were used to produce the model. It's still going to be computationally very expensive. Also, probably architecture of the hidden layers wouldn't be optimal.

My guess it that filtering on your own is just what you need.

Lukasz Tracewski
  • 10,794
  • 3
  • 34
  • 53
  • The model that I'm using already detects Spatulas with pretty high accuracy. Will retraining improve upon that or start anew? – legolasshegolas Oct 28 '16 at 07:26
  • Yes, it will make it better, if used correctly - a truly dedicated spatula detector. Mind that it might be not worth the effort. Refining (building upon existing network) will make it better. Retraining can give even better results, but it will require significantly more computational resources. – Lukasz Tracewski Oct 29 '16 at 11:49
  • @legolasshegolas Does this answer your question? In all of the questions you posted on SO you have not marked the correct answer. Are you sure none of the users provided you with a good answer? – Lukasz Tracewski Nov 05 '16 at 12:14