Bootstrapping Deep Neural Networks from Approximate Image Processing Pipelines


== Abstract ==

Complex image processing and computer vision systems often consist of a processing pipeline of functional modules. We intend to replace parts or all of a target pipeline with deep neural networks to achieve benefits such as increased accuracy or reduced computational requirement. To acquire a large amount of labeled data necessary to train the deep neural network, we propose a workflow that leverages the target pipeline to create a significantly larger labeled training set automatically, without prior domain knowledge of the target pipeline. We show experimentally that despite the noise introduced by automated labeling and only using a very small initially labeled data set, the trained deep neural networks can achieve similar or even better performance than the components they replace, while in some cases also reducing computational requirements.

Comment: 6 pages, 5 figure


== Original document ==

The different versions of the original document can be found in:

* [http://arxiv.org/abs/1811.12108 http://arxiv.org/abs/1811.12108]

* [http://arxiv.org/pdf/1811.12108 http://arxiv.org/pdf/1811.12108]

* [http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=9027223 http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=9027223],
: [http://xplorestaging.ieee.org/ielx7/9022708/9027210/09027223.pdf?arnumber=9027223 http://xplorestaging.ieee.org/ielx7/9022708/9027210/09027223.pdf?arnumber=9027223],
: [https://academic.microsoft.com/#/detail/3011172869 https://academic.microsoft.com/#/detail/3011172869]

* [http://xplorestaging.ieee.org/ielx7/9022708/9027210/09027223.pdf?arnumber=9027223 http://xplorestaging.ieee.org/ielx7/9022708/9027210/09027223.pdf?arnumber=9027223],
: [http://dx.doi.org/10.1109/emc249363.2019.00009 http://dx.doi.org/10.1109/emc249363.2019.00009]