This commit contains all significant contributions until 23/8/2016 #761
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Implemented ImagePreprocessor a geneology of functors for preprocessing samples before they are fed to a CNN. Implemented TextImageClassifer as an abstraction of any object that provides a vector of probabillities given an image. A gluing class providing compatibillity to callbacks for OCRBeamsearchDecoder allows to plug any such object to the existing algorithms. Implemented DeepCNN a class that is supposed to load pretrained caffe models. Although several backends have been planed the only fully ope- rational one is the one employing caffe its self. GPU is supported. Implemented a class using a DeepCNN that implements word-spotting Implemented a python and a C++ demo demonstraiting the WorSpotting Class Implemented a class that generates sythetic text samples at high speeds and with a full pipeline of realistic distortions for training Word-spotting CNN's who need millios of samples. Implemented a python demo demonstraiting the usage of the text synthesizer
…y to the synthesizer with the simple highgui
…keLists.txt substituting configure_file with add_definitions.
… empty file as was originally in the master branch
mshabunin
reviewed
Nov 22, 2016
| @@ -1,7 +1,13 @@ | |||
| #ifndef __OPENCV_TEXT_CONFIG_HPP__ | |||
| #define __OPENCV_TEXT_CONFIG_HPP__ | |||
|
|
|||
| // HAVE QT5 | |||
There was a problem hiding this comment.
Actually this file is not usable, because there is no corresponding configure_file command in cmake script (like here).
HAVE_TESSERACT works because the definition is being added to the compilation command (with add_definitions(-DHAVE_TESSERACT)).
Actually I think we should use the dnn module to run the network, is it possible?
| @@ -61,82 +65,126 @@ enum | |||
| OCR_LEVEL_TEXTLINE | |||
| }; | |||
|
|
|||
| //base class BaseOCR declares a common API that would be used in a typical text recognition scenario | |||
| //base class BaseOCR declares a common API that would be used in a typical text | |||
There was a problem hiding this comment.
Please, restore the formatting of all documentation comments and code.
| @@ -43,6 +43,7 @@ | |||
| #include "precomp.hpp" | |||
| #include "opencv2/imgproc.hpp" | |||
| #include "opencv2/ml.hpp" | |||
| #include "opencv2/highgui.hpp" | |||
| message(STATUS "TEXT CAFFE CONFLICT") | ||
| endif() | ||
|
|
||
| find_package(Qt5Gui) |
There was a problem hiding this comment.
It will be better to use HAVE_QT5 from the main repository. (examples)
|
Ping @anguelos . |
This was referenced Aug 28, 2017
Closed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This pullrequest changes
Implemented ImagePreprocessor a genealogy of functors for preprocessing
samples before they are fed to a CNN.
Implemented TextImageClassifer as an abstraction of any object that
provides a vector of probabilities given an image. A gluing class
providing compatibility to callbacks for OCRBeamsearchDecoder allows
to plug any such object to the existing algorithms.
Implemented DeepCNN a class that is supposed to load pre-trained Caffe
models. Although several back-ends have been planed the only operational
one is the one employing Caffe its self. GPU is fully supported but can only
be enabled and disabled globally.
Implemented a class using a DeepCNN that implements word-spotting
Implemented a python and a C++ demo demonstrating the Word-spotting
Class
Implemented a class that generates synthetic text samples at high speeds
and with a full pipeline of realistic distortions for training
Word-spotting CNN's who need millions of samples.
Implemented a python demo demonstrating the usage of the text synthesizer