File size: 39,178 Bytes
9375c9a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 |
<html><!-- Created using the cpp_pretty_printer from the dlib C++ library. See http://dlib.net for updates. --><head><title>dlib C++ Library - input_abstract.h</title></head><body bgcolor='white'><pre>
<font color='#009900'>// Copyright (C) 2015 Davis E. King ([email protected])
</font><font color='#009900'>// License: Boost Software License See LICENSE.txt for the full license.
</font><font color='#0000FF'>#undef</font> DLIB_DNn_INPUT_ABSTRACT_H_
<font color='#0000FF'>#ifdef</font> DLIB_DNn_INPUT_ABSTRACT_H_
<font color='#0000FF'>#include</font> "<a style='text-decoration:none' href='../matrix.h.html'>../matrix.h</a>"
<font color='#0000FF'>#include</font> "<a style='text-decoration:none' href='../pixel.h.html'>../pixel.h</a>"
<font color='#0000FF'>namespace</font> dlib
<b>{</b>
<font color='#009900'>// ----------------------------------------------------------------------------------------
</font>
<font color='#0000FF'>class</font> <b><a name='EXAMPLE_INPUT_LAYER'></a>EXAMPLE_INPUT_LAYER</b>
<b>{</b>
<font color='#009900'>/*!
WHAT THIS OBJECT REPRESENTS
Each deep neural network model in dlib begins with an input layer. The job
of the input layer is to convert an input_type into a tensor. Nothing more
and nothing less.
Note that there is no dlib::EXAMPLE_INPUT_LAYER type. It is shown here
purely to document the interface that an input layer object must implement.
If you are using some kind of image or matrix object as your input_type
then you can use the provided dlib::input layer defined below. Otherwise,
you need to define your own custom input layer.
THREAD SAFETY
to_tensor() must be thread safe. That is, multiple threads must be able to
make calls to to_tensor() on a single instance of this object at the same
time.
!*/</font>
<font color='#0000FF'>public</font>:
<b><a name='EXAMPLE_INPUT_LAYER'></a>EXAMPLE_INPUT_LAYER</b><font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- Default constructs this object. This function is not required to do
anything in particular but it must exist, that is, it is required that
layer objects be default constructable.
!*/</font>
<b><a name='EXAMPLE_INPUT_LAYER'></a>EXAMPLE_INPUT_LAYER</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'>const</font> EXAMPLE_INPUT_LAYER<font color='#5555FF'>&</font> item
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- EXAMPLE_INPUT_LAYER objects are copy constructable
!*/</font>
<b><a name='EXAMPLE_INPUT_LAYER'></a>EXAMPLE_INPUT_LAYER</b><font face='Lucida Console'>(</font>
<font color='#0000FF'>const</font> some_other_input_layer_type<font color='#5555FF'>&</font> item
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- Constructs this object from item. This form of constructor is optional
but it allows you to provide a conversion from one input layer type to
another. For example, the following code is valid only if my_input_layer2 can
be constructed from my_input_layer1:
relu<fc<relu<fc<my_input_layer1>>>> my_dnn1;
relu<fc<relu<fc<my_input_layer2>>>> my_dnn2(my_dnn1);
This kind of pattern is useful if you want to use one type of input layer
during training but a different type of layer during testing since it
allows you to easily convert between related deep neural network types.
!*/</font>
<font color='#0000FF'>typedef</font> whatever_type_to_tensor_expects input_type;
<font color='#0000FF'>template</font> <font color='#5555FF'><</font><font color='#0000FF'>typename</font> forward_iterator<font color='#5555FF'>></font>
<font color='#0000FF'><u>void</u></font> <b><a name='to_tensor'></a>to_tensor</b> <font face='Lucida Console'>(</font>
forward_iterator ibegin,
forward_iterator iend,
resizable_tensor<font color='#5555FF'>&</font> data
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- [ibegin, iend) is an iterator range over input_type objects.
- std::distance(ibegin,iend) > 0
ensures
- Converts the iterator range into a tensor and stores it into #data.
- #data.num_samples()%distance(ibegin,iend) == 0.
Normally you would have #data.num_samples() == distance(ibegin,iend) but
you can also expand the output by some integer factor so long as the loss
you use can deal with it correctly.
- The data in the ith sample of #data corresponds to the input_type object
*(ibegin+i/sample_expansion_factor).
where sample_expansion_factor==#data.num_samples()/distance(ibegin,iend).
!*/</font>
<b>}</b>;
std::ostream<font color='#5555FF'>&</font> <b><a name='operator'></a>operator</b><font color='#5555FF'><</font><font color='#5555FF'><</font><font face='Lucida Console'>(</font>std::ostream<font color='#5555FF'>&</font> out, <font color='#0000FF'>const</font> EXAMPLE_INPUT_LAYER<font color='#5555FF'>&</font> item<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
print a string describing this layer.
!*/</font>
<font color='#0000FF'><u>void</u></font> <b><a name='to_xml'></a>to_xml</b><font face='Lucida Console'>(</font><font color='#0000FF'>const</font> EXAMPLE_INPUT_LAYER<font color='#5555FF'>&</font> item, std::ostream<font color='#5555FF'>&</font> out<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
This function is optional, but required if you want to print your networks with
net_to_xml(). Therefore, to_xml() prints a layer as XML.
!*/</font>
<font color='#0000FF'><u>void</u></font> <b><a name='serialize'></a>serialize</b><font face='Lucida Console'>(</font><font color='#0000FF'>const</font> EXAMPLE_INPUT_LAYER<font color='#5555FF'>&</font> item, std::ostream<font color='#5555FF'>&</font> out<font face='Lucida Console'>)</font>;
<font color='#0000FF'><u>void</u></font> <b><a name='deserialize'></a>deserialize</b><font face='Lucida Console'>(</font>EXAMPLE_INPUT_LAYER<font color='#5555FF'>&</font> item, std::istream<font color='#5555FF'>&</font> in<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
provides serialization support
!*/</font>
<font color='#009900'>// ----------------------------------------------------------------------------------------
</font>
<font color='#0000FF'>template</font> <font color='#5555FF'><</font>
<font color='#0000FF'>typename</font> T
<font color='#5555FF'>></font>
<font color='#0000FF'>class</font> <b><a name='input'></a>input</b>
<b>{</b>
<font color='#009900'>/*!
REQUIREMENTS ON T
One of the following must be true:
- T is a matrix or array2d object and it must contain some kind of
pixel type. I.e. pixel_traits<T::type> must be defined.
- T is a std::array<matrix<U>> where U is any built in scalar type like
float, double, or unsigned char.
WHAT THIS OBJECT REPRESENTS
This is a basic input layer that simply copies images into a tensor.
!*/</font>
<font color='#0000FF'>public</font>:
<font color='#0000FF'>typedef</font> T input_type;
<font color='#0000FF'>template</font> <font color='#5555FF'><</font><font color='#0000FF'>typename</font> forward_iterator<font color='#5555FF'>></font>
<font color='#0000FF'><u>void</u></font> <b><a name='to_tensor'></a>to_tensor</b> <font face='Lucida Console'>(</font>
forward_iterator ibegin,
forward_iterator iend,
resizable_tensor<font color='#5555FF'>&</font> data
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- [ibegin, iend) is an iterator range over input_type objects.
- std::distance(ibegin,iend) > 0
- The input range should contain image objects that all have the same
dimensions.
ensures
- Converts the iterator range into a tensor and stores it into #data. In
particular, if the input images have R rows, C columns, and K channels
(where K is given by pixel_traits::num or std::array::size() if
std::array inputs are used) then we will have:
- #data.num_samples() == std::distance(ibegin,iend)
- #data.nr() == R
- #data.nc() == C
- #data.k() == K
For example, a matrix<float,3,3> would turn into a tensor with 3 rows, 3
columns, and k()==1. Or a matrix<rgb_pixel,4,5> would turn into a tensor
with 4 rows, 5 columns, and k()==3 (since rgb_pixels have 3 channels).
Or a std::array<matrix<float,3,3>,5> would turn into a tensor with 3 rows
and columns, and k()==5 channels.
- If the input data contains pixels of type unsigned char, rgb_pixel, or
other pixel types with a basic_pixel_type of unsigned char then each
value written to the output tensor is first divided by 256.0 so that the
resulting outputs are all in the range [0,1].
!*/</font>
<font color='#009900'>// Provided for compatibility with input_rgb_image_pyramid's interface
</font> <font color='#0000FF'><u>bool</u></font> <b><a name='image_contained_point'></a>image_contained_point</b> <font face='Lucida Console'>(</font> <font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> data, <font color='#0000FF'>const</font> point<font color='#5555FF'>&</font> p<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font> <b>{</b> <font color='#0000FF'>return</font> <font color='#BB00BB'>get_rect</font><font face='Lucida Console'>(</font>data<font face='Lucida Console'>)</font>.<font color='#BB00BB'>contains</font><font face='Lucida Console'>(</font>p<font face='Lucida Console'>)</font>; <b>}</b>
drectangle <b><a name='tensor_space_to_image_space'></a>tensor_space_to_image_space</b> <font face='Lucida Console'>(</font> <font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> <font color='#009900'>/*data*/</font>, drectangle r<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font> <b>{</b> <font color='#0000FF'>return</font> r; <b>}</b>
drectangle <b><a name='image_space_to_tensor_space'></a>image_space_to_tensor_space</b> <font face='Lucida Console'>(</font> <font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> <font color='#009900'>/*data*/</font>, <font color='#0000FF'><u>double</u></font> <font color='#009900'>/*scale*/</font>, drectangle r <font face='Lucida Console'>)</font> <font color='#0000FF'>const</font> <b>{</b> <font color='#0000FF'>return</font> r; <b>}</b>
<b>}</b>;
<font color='#009900'>// ----------------------------------------------------------------------------------------
</font>
<font color='#0000FF'>class</font> <b><a name='input_rgb_image'></a>input_rgb_image</b>
<b>{</b>
<font color='#009900'>/*!
WHAT THIS OBJECT REPRESENTS
This input layer works with RGB images of type matrix<rgb_pixel>. It is
very similar to the dlib::input layer except that it allows you to subtract
the average color value from each color channel when converting an image to
a tensor.
!*/</font>
<font color='#0000FF'>public</font>:
<font color='#0000FF'>typedef</font> matrix<font color='#5555FF'><</font>rgb_pixel<font color='#5555FF'>></font> input_type;
<b><a name='input_rgb_image'></a>input_rgb_image</b> <font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- #get_avg_red() == 122.782
- #get_avg_green() == 117.001
- #get_avg_blue() == 104.298
!*/</font>
<b><a name='input_rgb_image'></a>input_rgb_image</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'><u>float</u></font> avg_red,
<font color='#0000FF'><u>float</u></font> avg_green,
<font color='#0000FF'><u>float</u></font> avg_blue
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- #get_avg_red() == avg_red
- #get_avg_green() == avg_green
- #get_avg_blue() == avg_blue
!*/</font>
<font color='#0000FF'><u>float</u></font> <b><a name='get_avg_red'></a>get_avg_red</b><font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- returns the value subtracted from the red color channel.
!*/</font>
<font color='#0000FF'><u>float</u></font> <b><a name='get_avg_green'></a>get_avg_green</b><font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- returns the value subtracted from the green color channel.
!*/</font>
<font color='#0000FF'><u>float</u></font> <b><a name='get_avg_blue'></a>get_avg_blue</b><font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- returns the value subtracted from the blue color channel.
!*/</font>
<font color='#0000FF'>template</font> <font color='#5555FF'><</font><font color='#0000FF'>typename</font> forward_iterator<font color='#5555FF'>></font>
<font color='#0000FF'><u>void</u></font> <b><a name='to_tensor'></a>to_tensor</b> <font face='Lucida Console'>(</font>
forward_iterator ibegin,
forward_iterator iend,
resizable_tensor<font color='#5555FF'>&</font> data
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- [ibegin, iend) is an iterator range over input_type objects.
- std::distance(ibegin,iend) > 0
- The input range should contain images that all have the same
dimensions.
ensures
- Converts the iterator range into a tensor and stores it into #data. In
particular, if the input images have R rows, C columns then we will have:
- #data.num_samples() == std::distance(ibegin,iend)
- #data.nr() == R
- #data.nc() == C
- #data.k() == 3
Moreover, each color channel is normalized by having its average value
subtracted (according to get_avg_red(), get_avg_green(), or
get_avg_blue()) and then is divided by 256.0.
!*/</font>
<font color='#009900'>// Provided for compatibility with input_rgb_image_pyramid's interface
</font> <font color='#0000FF'><u>bool</u></font> <b><a name='image_contained_point'></a>image_contained_point</b> <font face='Lucida Console'>(</font> <font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> data, <font color='#0000FF'>const</font> point<font color='#5555FF'>&</font> p<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font> <b>{</b> <font color='#0000FF'>return</font> <font color='#BB00BB'>get_rect</font><font face='Lucida Console'>(</font>data<font face='Lucida Console'>)</font>.<font color='#BB00BB'>contains</font><font face='Lucida Console'>(</font>p<font face='Lucida Console'>)</font>; <b>}</b>
drectangle <b><a name='tensor_space_to_image_space'></a>tensor_space_to_image_space</b> <font face='Lucida Console'>(</font> <font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> <font color='#009900'>/*data*/</font>, drectangle r<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font> <b>{</b> <font color='#0000FF'>return</font> r; <b>}</b>
drectangle <b><a name='image_space_to_tensor_space'></a>image_space_to_tensor_space</b> <font face='Lucida Console'>(</font> <font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> <font color='#009900'>/*data*/</font>, <font color='#0000FF'><u>double</u></font> <font color='#009900'>/*scale*/</font>, drectangle r <font face='Lucida Console'>)</font> <font color='#0000FF'>const</font> <b>{</b> <font color='#0000FF'>return</font> r; <b>}</b>
<b>}</b>;
<font color='#009900'>// ----------------------------------------------------------------------------------------
</font>
<font color='#0000FF'>template</font> <font color='#5555FF'><</font><font color='#0000FF'><u>size_t</u></font> NR, <font color='#0000FF'><u>size_t</u></font> NC<font color='#5555FF'>=</font>NR<font color='#5555FF'>></font>
<font color='#0000FF'>class</font> <b><a name='input_rgb_image_sized'></a>input_rgb_image_sized</b>
<b>{</b>
<font color='#009900'>/*!
WHAT THIS OBJECT REPRESENTS
This layer has an interface and behavior identical to input_rgb_image
except that it requires input images to have NR rows and NC columns. This
is checked by a DLIB_CASSERT inside to_tensor().
You can also convert between input_rgb_image and input_rgb_image_sized by
copy construction or assignment.
!*/</font>
<b>}</b>;
<font color='#009900'>// ----------------------------------------------------------------------------------------
</font>
<font color='#0000FF'>template</font> <font color='#5555FF'><</font>
<font color='#0000FF'>typename</font> PYRAMID_TYPE
<font color='#5555FF'>></font>
<font color='#0000FF'>class</font> <b><a name='input_grayscale_image_pyramid'></a>input_grayscale_image_pyramid</b>
<b>{</b>
<font color='#009900'>/*!
REQUIREMENTS ON PYRAMID_TYPE
PYRAMID_TYPE must be an instance of the dlib::pyramid_down template.
WHAT THIS OBJECT REPRESENTS
This input layer works with gray scale images of type matrix<unsigned char>.
It is identical to input layer except that it outputs a tensor containing a tiled
image pyramid of each input image rather than a simple copy of each image.
The tiled image pyramid is created using create_tiled_pyramid().
!*/</font>
<font color='#0000FF'>public</font>:
<font color='#0000FF'>typedef</font> matrix<font color='#5555FF'><</font><font color='#0000FF'><u>unsigned</u></font> <font color='#0000FF'><u>char</u></font><font color='#5555FF'>></font> input_type;
<font color='#0000FF'>typedef</font> PYRAMID_TYPE pyramid_type;
<b><a name='input_grayscale_image_pyramid'></a>input_grayscale_image_pyramid</b> <font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- #get_pyramid_padding() == 10
- #get_pyramid_outer_padding() == 11
!*/</font>
<font color='#0000FF'><u>unsigned</u></font> <font color='#0000FF'><u>long</u></font> <b><a name='get_pyramid_padding'></a>get_pyramid_padding</b> <font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- When this object creates a pyramid it will call create_tiled_pyramid() and
set create_tiled_pyramid's pyramid_padding parameter to get_pyramid_padding().
!*/</font>
<font color='#0000FF'><u>void</u></font> <b><a name='set_pyramid_padding'></a>set_pyramid_padding</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'><u>unsigned</u></font> <font color='#0000FF'><u>long</u></font> value
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- #get_pyramid_padding() == value
!*/</font>
<font color='#0000FF'><u>unsigned</u></font> <font color='#0000FF'><u>long</u></font> <b><a name='get_pyramid_outer_padding'></a>get_pyramid_outer_padding</b> <font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- When this object creates a pyramid it will call create_tiled_pyramid()
and set create_tiled_pyramid's pyramid_outer_padding parameter to
get_pyramid_outer_padding().
!*/</font>
<font color='#0000FF'><u>void</u></font> <b><a name='set_pyramid_outer_padding'></a>set_pyramid_outer_padding</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'><u>unsigned</u></font> <font color='#0000FF'><u>long</u></font> value
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- #get_pyramid_outer_padding() == value
!*/</font>
<font color='#0000FF'>template</font> <font color='#5555FF'><</font><font color='#0000FF'>typename</font> forward_iterator<font color='#5555FF'>></font>
<font color='#0000FF'><u>void</u></font> <b><a name='to_tensor'></a>to_tensor</b> <font face='Lucida Console'>(</font>
forward_iterator ibegin,
forward_iterator iend,
resizable_tensor<font color='#5555FF'>&</font> data
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- [ibegin, iend) is an iterator range over input_type objects.
- std::distance(ibegin,iend) > 0
- The input range should contain images that all have the same
dimensions.
ensures
- Converts the iterator range into a tensor and stores it into #data. In
particular, we will have:
- #data.num_samples() == std::distance(ibegin,iend)
- #data.k() == 1
- Each sample in #data contains a tiled image pyramid of the
corresponding input image. The tiled pyramid is created by
create_tiled_pyramid().
Moreover, each pixel is normalized, dividing them by 256.0.
!*/</font>
<font color='#0000FF'><u>bool</u></font> <b><a name='image_contained_point'></a>image_contained_point</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> data,
<font color='#0000FF'>const</font> point<font color='#5555FF'>&</font> p
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- data is a tensor that was produced by this->to_tensor()
ensures
- Since data is a tensor that is built from a bunch of identically sized
images, we can ask if those images were big enough to contain the point
p. This function returns the answer to that question.
!*/</font>
drectangle <b><a name='image_space_to_tensor_space'></a>image_space_to_tensor_space</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> data,
<font color='#0000FF'><u>double</u></font> scale,
drectangle r
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- data is a tensor that was produced by this->to_tensor()
- 0 < scale <= 1
ensures
- This function maps from to_tensor()'s input image space to its output
tensor space. Therefore, given that data is a tensor produced by
to_tensor(), image_space_to_tensor_space() allows you to ask for the
rectangle in data that corresponds to a rectangle in the original image
space.
Note that since the output tensor contains an image pyramid, there are
multiple points in the output tensor that correspond to any input
location. So you must also specify a scale so we know what level of the
pyramid is needed. So given a rectangle r in an input image, you can
ask, what rectangle in data corresponds to r when things are scale times
smaller? That rectangle is returned by this function.
- A scale of 1 means we don't move anywhere in the pyramid scale space relative
to the input image while smaller values of scale mean we move down the
pyramid.
!*/</font>
drectangle <b><a name='tensor_space_to_image_space'></a>tensor_space_to_image_space</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> data,
drectangle r
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- data is a tensor that was produced by this->to_tensor()
ensures
- This function maps from to_tensor()'s output tensor space to its input
image space. Therefore, given that data is a tensor produced by
to_tensor(), tensor_space_to_image_space() allows you to ask for the
rectangle in the input image that corresponds to a rectangle in data.
- It should be noted that this function isn't always an inverse of
image_space_to_tensor_space(). This is because you can ask
image_space_to_tensor_space() for the coordinates of points outside the input
image and they will be mapped to somewhere that doesn't have an inverse.
But for points actually inside the input image this function performs an
approximate inverse mapping. I.e. when image_contained_point(data,center(r))==true
there is an approximate inverse.
!*/</font>
<b>}</b>;
<font color='#009900'>// ----------------------------------------------------------------------------------------
</font>
<font color='#0000FF'>template</font> <font color='#5555FF'><</font>
<font color='#0000FF'>typename</font> PYRAMID_TYPE
<font color='#5555FF'>></font>
<font color='#0000FF'>class</font> <b><a name='input_rgb_image_pyramid'></a>input_rgb_image_pyramid</b>
<b>{</b>
<font color='#009900'>/*!
REQUIREMENTS ON PYRAMID_TYPE
PYRAMID_TYPE must be an instance of the dlib::pyramid_down template.
WHAT THIS OBJECT REPRESENTS
This input layer works with RGB images of type matrix<rgb_pixel>. It is
identical to input_rgb_image except that it outputs a tensor containing a
tiled image pyramid of each input image rather than a simple copy of each
image. The tiled image pyramid is created using create_tiled_pyramid().
!*/</font>
<font color='#0000FF'>public</font>:
<font color='#0000FF'>typedef</font> matrix<font color='#5555FF'><</font>rgb_pixel<font color='#5555FF'>></font> input_type;
<font color='#0000FF'>typedef</font> PYRAMID_TYPE pyramid_type;
<b><a name='input_rgb_image_pyramid'></a>input_rgb_image_pyramid</b> <font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- #get_avg_red() == 122.782
- #get_avg_green() == 117.001
- #get_avg_blue() == 104.298
- #get_pyramid_padding() == 10
- #get_pyramid_outer_padding() == 11
!*/</font>
<b><a name='input_rgb_image_pyramid'></a>input_rgb_image_pyramid</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'><u>float</u></font> avg_red,
<font color='#0000FF'><u>float</u></font> avg_green,
<font color='#0000FF'><u>float</u></font> avg_blue
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- #get_avg_red() == avg_red
- #get_avg_green() == avg_green
- #get_avg_blue() == avg_blue
- #get_pyramid_padding() == 10
- #get_pyramid_outer_padding() == 11
!*/</font>
<font color='#0000FF'><u>float</u></font> <b><a name='get_avg_red'></a>get_avg_red</b><font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- returns the value subtracted from the red color channel.
!*/</font>
<font color='#0000FF'><u>float</u></font> <b><a name='get_avg_green'></a>get_avg_green</b><font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- returns the value subtracted from the green color channel.
!*/</font>
<font color='#0000FF'><u>float</u></font> <b><a name='get_avg_blue'></a>get_avg_blue</b><font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- returns the value subtracted from the blue color channel.
!*/</font>
<font color='#0000FF'><u>unsigned</u></font> <font color='#0000FF'><u>long</u></font> <b><a name='get_pyramid_padding'></a>get_pyramid_padding</b> <font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- When this object creates a pyramid it will call create_tiled_pyramid() and
set create_tiled_pyramid's pyramid_padding parameter to get_pyramid_padding().
!*/</font>
<font color='#0000FF'><u>void</u></font> <b><a name='set_pyramid_padding'></a>set_pyramid_padding</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'><u>unsigned</u></font> <font color='#0000FF'><u>long</u></font> value
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- #get_pyramid_padding() == value
!*/</font>
<font color='#0000FF'><u>unsigned</u></font> <font color='#0000FF'><u>long</u></font> <b><a name='get_pyramid_outer_padding'></a>get_pyramid_outer_padding</b> <font face='Lucida Console'>(</font>
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
ensures
- When this object creates a pyramid it will call create_tiled_pyramid()
and set create_tiled_pyramid's pyramid_outer_padding parameter to
get_pyramid_outer_padding().
!*/</font>
<font color='#0000FF'><u>void</u></font> <b><a name='set_pyramid_outer_padding'></a>set_pyramid_outer_padding</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'><u>unsigned</u></font> <font color='#0000FF'><u>long</u></font> value
<font face='Lucida Console'>)</font>;
<font color='#009900'>/*!
ensures
- #get_pyramid_outer_padding() == value
!*/</font>
<font color='#0000FF'>template</font> <font color='#5555FF'><</font><font color='#0000FF'>typename</font> forward_iterator<font color='#5555FF'>></font>
<font color='#0000FF'><u>void</u></font> <b><a name='to_tensor'></a>to_tensor</b> <font face='Lucida Console'>(</font>
forward_iterator ibegin,
forward_iterator iend,
resizable_tensor<font color='#5555FF'>&</font> data
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- [ibegin, iend) is an iterator range over input_type objects.
- std::distance(ibegin,iend) > 0
- The input range should contain images that all have the same
dimensions.
ensures
- Converts the iterator range into a tensor and stores it into #data. In
particular, we will have:
- #data.num_samples() == std::distance(ibegin,iend)
- #data.k() == 3
- Each sample in #data contains a tiled image pyramid of the
corresponding input image. The tiled pyramid is created by
create_tiled_pyramid().
Moreover, each color channel is normalized by having its average value
subtracted (according to get_avg_red(), get_avg_green(), or
get_avg_blue()) and then is divided by 256.0.
!*/</font>
<font color='#0000FF'><u>bool</u></font> <b><a name='image_contained_point'></a>image_contained_point</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> data,
<font color='#0000FF'>const</font> point<font color='#5555FF'>&</font> p
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- data is a tensor that was produced by this->to_tensor()
ensures
- Since data is a tensor that is built from a bunch of identically sized
images, we can ask if those images were big enough to contain the point
p. This function returns the answer to that question.
!*/</font>
drectangle <b><a name='image_space_to_tensor_space'></a>image_space_to_tensor_space</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> data,
<font color='#0000FF'><u>double</u></font> scale,
drectangle r
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- data is a tensor that was produced by this->to_tensor()
- 0 < scale <= 1
ensures
- This function maps from to_tensor()'s input image space to its output
tensor space. Therefore, given that data is a tensor produced by
to_tensor(), image_space_to_tensor_space() allows you to ask for the
rectangle in data that corresponds to a rectangle in the original image
space.
Note that since the output tensor contains an image pyramid, there are
multiple points in the output tensor that correspond to any input
location. So you must also specify a scale so we know what level of the
pyramid is needed. So given a rectangle r in an input image, you can
ask, what rectangle in data corresponds to r when things are scale times
smaller? That rectangle is returned by this function.
- A scale of 1 means we don't move anywhere in the pyramid scale space relative
to the input image while smaller values of scale mean we move down the
pyramid.
!*/</font>
drectangle <b><a name='tensor_space_to_image_space'></a>tensor_space_to_image_space</b> <font face='Lucida Console'>(</font>
<font color='#0000FF'>const</font> tensor<font color='#5555FF'>&</font> data,
drectangle r
<font face='Lucida Console'>)</font> <font color='#0000FF'>const</font>;
<font color='#009900'>/*!
requires
- data is a tensor that was produced by this->to_tensor()
ensures
- This function maps from to_tensor()'s output tensor space to its input
image space. Therefore, given that data is a tensor produced by
to_tensor(), tensor_space_to_image_space() allows you to ask for the
rectangle in the input image that corresponds to a rectangle in data.
- It should be noted that this function isn't always an inverse of
image_space_to_tensor_space(). This is because you can ask
image_space_to_tensor_space() for the coordinates of points outside the input
image and they will be mapped to somewhere that doesn't have an inverse.
But for points actually inside the input image this function performs an
approximate inverse mapping. I.e. when image_contained_point(data,center(r))==true
there is an approximate inverse.
!*/</font>
<b>}</b>;
<font color='#009900'>// ----------------------------------------------------------------------------------------
</font>
<b>}</b>
<font color='#0000FF'>#endif</font> <font color='#009900'>// DLIB_DNn_INPUT_ABSTRACT_H_
</font>
</pre></body></html> |