Spaces:

AshanGimhana
/

Aging_MouthReplace

Paused

App Files Files Community

Aging_MouthReplace / dlibs /docs /dnn_instance_segmentation_ex.cpp.html

AshanGimhana

Upload folder using huggingface_hub

9375c9a verified 2 months ago

raw

history blame contribute delete

23.3 kB

	<html><!-- Created using the cpp_pretty_printer from the dlib C++ library. See http://dlib.net for updates. --><head><title>dlib C++ Library - dnn_instance_segmentation_ex.cpp</title></head><body bgcolor='white'><pre>
	<font color='#009900'>// The contents of this file are in the public domain. See LICENSE_FOR_EXAMPLE_PROGRAMS.txt
	</font><font color='#009900'>/*
	This example shows how to do instance segmentation on an image using net pretrained
	on the PASCAL VOC2012 dataset. For an introduction to what instance segmentation is,
	see the accompanying header file dnn_instance_segmentation_ex.h.

	Instructions how to run the example:
	1. Download the PASCAL VOC2012 data, and untar it somewhere.
	http://host.robots.ox.ac.uk/pascal/VOC/voc2012/VOCtrainval_11-May-2012.tar
	2. Build the dnn_instance_segmentation_train_ex example program.
	3. Run:
	./dnn_instance_segmentation_train_ex /path/to/VOC2012
	4. Wait while the network is being trained.
	5. Build the dnn_instance_segmentation_ex example program.
	6. Run:
	./dnn_instance_segmentation_ex /path/to/VOC2012-or-other-images

	An alternative to steps 2-4 above is to download a pre-trained network
	from here: http://dlib.net/files/instance_segmentation_voc2012net_v2.dnn

	It would be a good idea to become familiar with dlib's DNN tooling before reading this
	example. So you should read <a href="dnn_introduction_ex.cpp.html">dnn_introduction_ex.cpp</a> and <a href="dnn_introduction2_ex.cpp.html">dnn_introduction2_ex.cpp</a>
	before reading this example program.
	*/</font>

	<font color='#0000FF'>#include</font> "<a style='text-decoration:none' href='dnn_instance_segmentation_ex.h.html'>dnn_instance_segmentation_ex.h</a>"
	<font color='#0000FF'>#include</font> "<a style='text-decoration:none' href='pascal_voc_2012.h.html'>pascal_voc_2012.h</a>"

	<font color='#0000FF'>#include</font> <font color='#5555FF'><</font>iostream<font color='#5555FF'>></font>
	<font color='#0000FF'>#include</font> <font color='#5555FF'><</font>dlib<font color='#5555FF'>/</font>data_io.h<font color='#5555FF'>></font>
	<font color='#0000FF'>#include</font> <font color='#5555FF'><</font>dlib<font color='#5555FF'>/</font>gui_widgets.h<font color='#5555FF'>></font>

	<font color='#0000FF'>using</font> <font color='#0000FF'>namespace</font> std;
	<font color='#0000FF'>using</font> <font color='#0000FF'>namespace</font> dlib;

	<font color='#009900'>// ----------------------------------------------------------------------------------------
	</font>
	<font color='#0000FF'><u>int</u></font> <b><a name='main'></a>main</b><font face='Lucida Console'>(</font><font color='#0000FF'><u>int</u></font> argc, <font color='#0000FF'><u>char</u></font><font color='#5555FF'></font><font color='#5555FF'></font> argv<font face='Lucida Console'>)</font> <font color='#0000FF'>try</font>
	<b>{</b>
	<font color='#0000FF'>if</font> <font face='Lucida Console'>(</font>argc <font color='#5555FF'>!</font><font color='#5555FF'>=</font> <font color='#979000'>2</font><font face='Lucida Console'>)</font>
	<b>{</b>
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>You call this program like this: </font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> endl;
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>./dnn_instance_segmentation_train_ex /path/to/images</font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> endl;
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> endl;
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>You will also need a trained '</font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> instance_segmentation_net_filename <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>' file.</font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> endl;
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>You can either train it yourself (see example program</font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> endl;
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>dnn_instance_segmentation_train_ex), or download a</font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> endl;
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>copy from here: http://dlib.net/files/</font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> instance_segmentation_net_filename <font color='#5555FF'><</font><font color='#5555FF'><</font> endl;
	<font color='#0000FF'>return</font> <font color='#979000'>1</font>;
	<b>}</b>

	<font color='#009900'>// Read the file containing the trained networks from the working directory.
	</font> det_anet_type det_net;
	std::map<font color='#5555FF'><</font>std::string, seg_bnet_type<font color='#5555FF'>></font> seg_nets_by_class;
	<font color='#BB00BB'>deserialize</font><font face='Lucida Console'>(</font>instance_segmentation_net_filename<font face='Lucida Console'>)</font> <font color='#5555FF'>></font><font color='#5555FF'>></font> det_net <font color='#5555FF'>></font><font color='#5555FF'>></font> seg_nets_by_class;

	<font color='#009900'>// Show inference results in a window.
	</font> image_window win;

	matrix<font color='#5555FF'><</font>rgb_pixel<font color='#5555FF'>></font> input_image;

	<font color='#009900'>// Find supported image files.
	</font> <font color='#0000FF'>const</font> std::vector<font color='#5555FF'><</font>file<font color='#5555FF'>></font> files <font color='#5555FF'>=</font> dlib::<font color='#BB00BB'>get_files_in_directory_tree</font><font face='Lucida Console'>(</font>argv[<font color='#979000'>1</font>],
	dlib::<font color='#BB00BB'>match_endings</font><font face='Lucida Console'>(</font>"<font color='#CC0000'>.jpeg .jpg .png</font>"<font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>;

	dlib::rand rnd;

	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>Found </font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> files.<font color='#BB00BB'>size</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font> <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'> images, processing...</font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> endl;

	<font color='#0000FF'>for</font> <font face='Lucida Console'>(</font><font color='#0000FF'>const</font> file<font color='#5555FF'>&</font> file : files<font face='Lucida Console'>)</font>
	<b>{</b>
	<font color='#009900'>// Load the input image.
	</font> <font color='#BB00BB'>load_image</font><font face='Lucida Console'>(</font>input_image, file.<font color='#BB00BB'>full_name</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>;

	<font color='#009900'>// Find instances in the input image
	</font> <font color='#0000FF'>const</font> <font color='#0000FF'>auto</font> instances <font color='#5555FF'>=</font> <font color='#BB00BB'>det_net</font><font face='Lucida Console'>(</font>input_image<font face='Lucida Console'>)</font>;

	matrix<font color='#5555FF'><</font>rgb_pixel<font color='#5555FF'>></font> rgb_label_image;
	matrix<font color='#5555FF'><</font><font color='#0000FF'><u>float</u></font><font color='#5555FF'>></font> label_image_confidence;

	matrix<font color='#5555FF'><</font>rgb_pixel<font color='#5555FF'>></font> input_chip;

	rgb_label_image.<font color='#BB00BB'>set_size</font><font face='Lucida Console'>(</font>input_image.<font color='#BB00BB'>nr</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font>, input_image.<font color='#BB00BB'>nc</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>;
	rgb_label_image <font color='#5555FF'>=</font> <font color='#BB00BB'>rgb_pixel</font><font face='Lucida Console'>(</font><font color='#979000'>0</font>, <font color='#979000'>0</font>, <font color='#979000'>0</font><font face='Lucida Console'>)</font>;

	label_image_confidence.<font color='#BB00BB'>set_size</font><font face='Lucida Console'>(</font>input_image.<font color='#BB00BB'>nr</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font>, input_image.<font color='#BB00BB'>nc</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>;
	label_image_confidence <font color='#5555FF'>=</font> <font color='#979000'>0.0</font>;

	<font color='#0000FF'><u>bool</u></font> found_something <font color='#5555FF'>=</font> <font color='#979000'>false</font>;

	<font color='#0000FF'>for</font> <font face='Lucida Console'>(</font><font color='#0000FF'>const</font> <font color='#0000FF'>auto</font><font color='#5555FF'>&</font> instance : instances<font face='Lucida Console'>)</font>
	<b>{</b>
	<font color='#0000FF'>if</font> <font face='Lucida Console'>(</font><font color='#5555FF'>!</font>found_something<font face='Lucida Console'>)</font>
	<b>{</b>
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>Found </font>";
	found_something <font color='#5555FF'>=</font> <font color='#979000'>true</font>;
	<b>}</b>
	<font color='#0000FF'>else</font>
	<b>{</b>
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'>, </font>";
	<b>}</b>
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> instance.label;

	<font color='#0000FF'>const</font> <font color='#0000FF'>auto</font> cropping_rect <font color='#5555FF'>=</font> <font color='#BB00BB'>get_cropping_rect</font><font face='Lucida Console'>(</font>instance.rect<font face='Lucida Console'>)</font>;
	<font color='#0000FF'>const</font> chip_details <font color='#BB00BB'>chip_details</font><font face='Lucida Console'>(</font>cropping_rect, <font color='#BB00BB'>chip_dims</font><font face='Lucida Console'>(</font>seg_dim, seg_dim<font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>;
	<font color='#BB00BB'>extract_image_chip</font><font face='Lucida Console'>(</font>input_image, chip_details, input_chip, <font color='#BB00BB'>interpolate_bilinear</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>;

	<font color='#0000FF'>const</font> <font color='#0000FF'>auto</font> i <font color='#5555FF'>=</font> seg_nets_by_class.<font color='#BB00BB'>find</font><font face='Lucida Console'>(</font>instance.label<font face='Lucida Console'>)</font>;
	<font color='#0000FF'>if</font> <font face='Lucida Console'>(</font>i <font color='#5555FF'>=</font><font color='#5555FF'>=</font> seg_nets_by_class.<font color='#BB00BB'>end</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>
	<b>{</b>
	<font color='#009900'>// per-class segmentation net not found, so we must be using the same net for all classes
	</font> <font color='#009900'>// (see bool separate_seg_net_for_each_class in <a href="dnn_instance_segmentation_train_ex.cpp.html">dnn_instance_segmentation_train_ex.cpp</a>)
	</font> <font color='#BB00BB'>DLIB_CASSERT</font><font face='Lucida Console'>(</font>seg_nets_by_class.<font color='#BB00BB'>size</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font> <font color='#5555FF'>=</font><font color='#5555FF'>=</font> <font color='#979000'>1</font><font face='Lucida Console'>)</font>;
	<font color='#BB00BB'>DLIB_CASSERT</font><font face='Lucida Console'>(</font>seg_nets_by_class.<font color='#BB00BB'>begin</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font color='#5555FF'>-</font><font color='#5555FF'>></font>first <font color='#5555FF'>=</font><font color='#5555FF'>=</font> "<font color='#CC0000'></font>"<font face='Lucida Console'>)</font>;
	<b>}</b>

	<font color='#0000FF'>auto</font><font color='#5555FF'>&</font> seg_net <font color='#5555FF'>=</font> i <font color='#5555FF'>!</font><font color='#5555FF'>=</font> seg_nets_by_class.<font color='#BB00BB'>end</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font>
	? i<font color='#5555FF'>-</font><font color='#5555FF'>></font>second <font color='#009900'>// use the segmentation net trained for this class
	</font> : seg_nets_by_class.<font color='#BB00BB'>begin</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font color='#5555FF'>-</font><font color='#5555FF'>></font>second; <font color='#009900'>// use the same segmentation net for all classes
	</font>
	<font color='#0000FF'>const</font> <font color='#0000FF'>auto</font> mask <font color='#5555FF'>=</font> <font color='#BB00BB'>seg_net</font><font face='Lucida Console'>(</font>input_chip<font face='Lucida Console'>)</font>;

	<font color='#0000FF'>const</font> rgb_pixel <font color='#BB00BB'>random_color</font><font face='Lucida Console'>(</font>
	rnd.<font color='#BB00BB'>get_random_8bit_number</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font>,
	rnd.<font color='#BB00BB'>get_random_8bit_number</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font>,
	rnd.<font color='#BB00BB'>get_random_8bit_number</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font>
	<font face='Lucida Console'>)</font>;

	dlib::matrix<font color='#5555FF'><</font><font color='#0000FF'><u>float</u></font><font color='#5555FF'>></font> <font color='#BB00BB'>resized_mask</font><font face='Lucida Console'>(</font>
	<font color='#0000FF'>static_cast</font><font color='#5555FF'><</font><font color='#0000FF'><u>int</u></font><font color='#5555FF'>></font><font face='Lucida Console'>(</font>chip_details.rect.<font color='#BB00BB'>height</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>,
	<font color='#0000FF'>static_cast</font><font color='#5555FF'><</font><font color='#0000FF'><u>int</u></font><font color='#5555FF'>></font><font face='Lucida Console'>(</font>chip_details.rect.<font color='#BB00BB'>width</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>
	<font face='Lucida Console'>)</font>;

	dlib::<font color='#BB00BB'>resize_image</font><font face='Lucida Console'>(</font>mask, resized_mask<font face='Lucida Console'>)</font>;

	<font color='#0000FF'>for</font> <font face='Lucida Console'>(</font><font color='#0000FF'><u>int</u></font> r <font color='#5555FF'>=</font> <font color='#979000'>0</font>; r <font color='#5555FF'><</font> resized_mask.<font color='#BB00BB'>nr</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font>; <font color='#5555FF'>+</font><font color='#5555FF'>+</font>r<font face='Lucida Console'>)</font>
	<b>{</b>
	<font color='#0000FF'>for</font> <font face='Lucida Console'>(</font><font color='#0000FF'><u>int</u></font> c <font color='#5555FF'>=</font> <font color='#979000'>0</font>; c <font color='#5555FF'><</font> resized_mask.<font color='#BB00BB'>nc</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font>; <font color='#5555FF'>+</font><font color='#5555FF'>+</font>c<font face='Lucida Console'>)</font>
	<b>{</b>
	<font color='#0000FF'>const</font> <font color='#0000FF'>auto</font> new_confidence <font color='#5555FF'>=</font> <font color='#BB00BB'>resized_mask</font><font face='Lucida Console'>(</font>r, c<font face='Lucida Console'>)</font>;
	<font color='#0000FF'>if</font> <font face='Lucida Console'>(</font>new_confidence <font color='#5555FF'>></font> <font color='#979000'>0</font><font face='Lucida Console'>)</font>
	<b>{</b>
	<font color='#0000FF'>const</font> <font color='#0000FF'>auto</font> y <font color='#5555FF'>=</font> chip_details.rect.<font color='#BB00BB'>top</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font> <font color='#5555FF'>+</font> r;
	<font color='#0000FF'>const</font> <font color='#0000FF'>auto</font> x <font color='#5555FF'>=</font> chip_details.rect.<font color='#BB00BB'>left</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font> <font color='#5555FF'>+</font> c;
	<font color='#0000FF'>if</font> <font face='Lucida Console'>(</font>y <font color='#5555FF'>></font><font color='#5555FF'>=</font> <font color='#979000'>0</font> <font color='#5555FF'>&</font><font color='#5555FF'>&</font> y <font color='#5555FF'><</font> rgb_label_image.<font color='#BB00BB'>nr</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font> <font color='#5555FF'>&</font><font color='#5555FF'>&</font> x <font color='#5555FF'>></font><font color='#5555FF'>=</font> <font color='#979000'>0</font> <font color='#5555FF'>&</font><font color='#5555FF'>&</font> x <font color='#5555FF'><</font> rgb_label_image.<font color='#BB00BB'>nc</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>
	<b>{</b>
	<font color='#0000FF'>auto</font><font color='#5555FF'>&</font> current_confidence <font color='#5555FF'>=</font> <font color='#BB00BB'>label_image_confidence</font><font face='Lucida Console'>(</font>y, x<font face='Lucida Console'>)</font>;
	<font color='#0000FF'>if</font> <font face='Lucida Console'>(</font>new_confidence <font color='#5555FF'>></font> current_confidence<font face='Lucida Console'>)</font>
	<b>{</b>
	<font color='#0000FF'>auto</font> rgb_label <font color='#5555FF'>=</font> random_color;
	<font color='#0000FF'>const</font> <font color='#0000FF'>auto</font> baseline_confidence <font color='#5555FF'>=</font> <font color='#979000'>5</font>;
	<font color='#0000FF'>if</font> <font face='Lucida Console'>(</font>new_confidence <font color='#5555FF'><</font> baseline_confidence<font face='Lucida Console'>)</font>
	<b>{</b>
	<font color='#009900'>// Scale label intensity if confidence isn't high
	</font> rgb_label.red <font color='#5555FF'>*</font><font color='#5555FF'>=</font> new_confidence <font color='#5555FF'>/</font> baseline_confidence;
	rgb_label.green <font color='#5555FF'>*</font><font color='#5555FF'>=</font> new_confidence <font color='#5555FF'>/</font> baseline_confidence;
	rgb_label.blue <font color='#5555FF'>*</font><font color='#5555FF'>=</font> new_confidence <font color='#5555FF'>/</font> baseline_confidence;
	<b>}</b>
	<font color='#BB00BB'>rgb_label_image</font><font face='Lucida Console'>(</font>y, x<font face='Lucida Console'>)</font> <font color='#5555FF'>=</font> rgb_label;
	current_confidence <font color='#5555FF'>=</font> new_confidence;
	<b>}</b>
	<b>}</b>
	<b>}</b>
	<b>}</b>
	<b>}</b>

	<font color='#0000FF'>const</font> Voc2012class<font color='#5555FF'>&</font> voc2012_class <font color='#5555FF'>=</font> <font color='#BB00BB'>find_voc2012_class</font><font face='Lucida Console'>(</font>
	[<font color='#5555FF'>&</font>instance]<font face='Lucida Console'>(</font><font color='#0000FF'>const</font> Voc2012class<font color='#5555FF'>&</font> candidate<font face='Lucida Console'>)</font> <b>{</b>
	<font color='#0000FF'>return</font> candidate.classlabel <font color='#5555FF'>=</font><font color='#5555FF'>=</font> instance.label;
	<b>}</b>
	<font face='Lucida Console'>)</font>;

	dlib::<font color='#BB00BB'>draw_rectangle</font><font face='Lucida Console'>(</font>rgb_label_image, instance.rect, voc2012_class.rgb_label, <font color='#979000'>1</font><font face='Lucida Console'>)</font>;
	<b>}</b>

	<font color='#009900'>// Show the input image on the left, and the predicted RGB labels on the right.
	</font> win.<font color='#BB00BB'>set_image</font><font face='Lucida Console'>(</font><font color='#BB00BB'>join_rows</font><font face='Lucida Console'>(</font>input_image, rgb_label_image<font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>;

	<font color='#0000FF'>if</font> <font face='Lucida Console'>(</font><font color='#5555FF'>!</font>instances.<font color='#BB00BB'>empty</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font><font face='Lucida Console'>)</font>
	<b>{</b>
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'> in </font>" <font color='#5555FF'><</font><font color='#5555FF'><</font> file.<font color='#BB00BB'>name</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font> <font color='#5555FF'><</font><font color='#5555FF'><</font> "<font color='#CC0000'> - hit enter to process the next image</font>";
	cin.<font color='#BB00BB'>get</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font>;
	<b>}</b>
	<b>}</b>
	<b>}</b>
	<font color='#0000FF'>catch</font><font face='Lucida Console'>(</font>std::exception<font color='#5555FF'>&</font> e<font face='Lucida Console'>)</font>
	<b>{</b>
	cout <font color='#5555FF'><</font><font color='#5555FF'><</font> e.<font color='#BB00BB'>what</font><font face='Lucida Console'>(</font><font face='Lucida Console'>)</font> <font color='#5555FF'><</font><font color='#5555FF'><</font> endl;
	<b>}</b>


	</pre></body></html>