vec_distance.hpp 5.4 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189
  1. /*M///////////////////////////////////////////////////////////////////////////////////////
  2. //
  3. // IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
  4. //
  5. // By downloading, copying, installing or using the software you agree to this license.
  6. // If you do not agree to this license, do not download, install,
  7. // copy or use the software.
  8. //
  9. //
  10. // License Agreement
  11. // For Open Source Computer Vision Library
  12. //
  13. // Copyright (C) 2000-2008, Intel Corporation, all rights reserved.
  14. // Copyright (C) 2009, Willow Garage Inc., all rights reserved.
  15. // Copyright (C) 2013, OpenCV Foundation, all rights reserved.
  16. // Third party copyrights are property of their respective owners.
  17. //
  18. // Redistribution and use in source and binary forms, with or without modification,
  19. // are permitted provided that the following conditions are met:
  20. //
  21. // * Redistribution's of source code must retain the above copyright notice,
  22. // this list of conditions and the following disclaimer.
  23. //
  24. // * Redistribution's in binary form must reproduce the above copyright notice,
  25. // this list of conditions and the following disclaimer in the documentation
  26. // and/or other materials provided with the distribution.
  27. //
  28. // * The name of the copyright holders may not be used to endorse or promote products
  29. // derived from this software without specific prior written permission.
  30. //
  31. // This software is provided by the copyright holders and contributors "as is" and
  32. // any express or implied warranties, including, but not limited to, the implied
  33. // warranties of merchantability and fitness for a particular purpose are disclaimed.
  34. // In no event shall the Intel Corporation or contributors be liable for any direct,
  35. // indirect, incidental, special, exemplary, or consequential damages
  36. // (including, but not limited to, procurement of substitute goods or services;
  37. // loss of use, data, or profits; or business interruption) however caused
  38. // and on any theory of liability, whether in contract, strict liability,
  39. // or tort (including negligence or otherwise) arising in any way out of
  40. // the use of this software, even if advised of the possibility of such damage.
  41. //
  42. //M*/
  43. #pragma once
  44. #ifndef OPENCV_CUDEV_BLOCK_VEC_DISTANCE_HPP
  45. #define OPENCV_CUDEV_BLOCK_VEC_DISTANCE_HPP
  46. #include "../common.hpp"
  47. #include "../functional/functional.hpp"
  48. #include "../warp/reduce.hpp"
  49. #include "reduce.hpp"
  50. namespace cv { namespace cudev {
  51. //! @addtogroup cudev
  52. //! @{
  53. // NormL1
  54. template <typename T> struct NormL1
  55. {
  56. typedef int value_type;
  57. typedef uint result_type;
  58. result_type mySum;
  59. __device__ __forceinline__ NormL1() : mySum(0) {}
  60. __device__ __forceinline__ void reduceThread(value_type val1, value_type val2)
  61. {
  62. mySum = __sad(val1, val2, mySum);
  63. }
  64. __device__ __forceinline__ void reduceWarp(result_type* smem, uint tid)
  65. {
  66. warpReduce(smem, mySum, tid, plus<result_type>());
  67. }
  68. template <int THREAD_DIM> __device__ __forceinline__ void reduceBlock(result_type* smem, uint tid)
  69. {
  70. blockReduce<THREAD_DIM>(smem, mySum, tid, plus<result_type>());
  71. }
  72. __device__ __forceinline__ operator result_type() const
  73. {
  74. return mySum;
  75. }
  76. };
  77. template <> struct NormL1<float>
  78. {
  79. typedef float value_type;
  80. typedef float result_type;
  81. result_type mySum;
  82. __device__ __forceinline__ NormL1() : mySum(0.0f) {}
  83. __device__ __forceinline__ void reduceThread(value_type val1, value_type val2)
  84. {
  85. mySum += ::fabsf(val1 - val2);
  86. }
  87. __device__ __forceinline__ void reduceWarp(result_type* smem, uint tid)
  88. {
  89. warpReduce(smem, mySum, tid, plus<result_type>());
  90. }
  91. template <int THREAD_DIM> __device__ __forceinline__ void reduceBlock(result_type* smem, uint tid)
  92. {
  93. blockReduce<THREAD_DIM>(smem, mySum, tid, plus<result_type>());
  94. }
  95. __device__ __forceinline__ operator result_type() const
  96. {
  97. return mySum;
  98. }
  99. };
  100. // NormL2
  101. struct NormL2
  102. {
  103. typedef float value_type;
  104. typedef float result_type;
  105. result_type mySum;
  106. __device__ __forceinline__ NormL2() : mySum(0.0f) {}
  107. __device__ __forceinline__ void reduceThread(value_type val1, value_type val2)
  108. {
  109. const float diff = val1 - val2;
  110. mySum += diff * diff;
  111. }
  112. __device__ __forceinline__ void reduceWarp(result_type* smem, uint tid)
  113. {
  114. warpReduce(smem, mySum, tid, plus<result_type>());
  115. }
  116. template <int THREAD_DIM> __device__ __forceinline__ void reduceBlock(result_type* smem, uint tid)
  117. {
  118. blockReduce<THREAD_DIM>(smem, mySum, tid, plus<result_type>());
  119. }
  120. __device__ __forceinline__ operator result_type() const
  121. {
  122. return ::sqrtf(mySum);
  123. }
  124. };
  125. // NormHamming
  126. struct NormHamming
  127. {
  128. typedef int value_type;
  129. typedef int result_type;
  130. result_type mySum;
  131. __device__ __forceinline__ NormHamming() : mySum(0) {}
  132. __device__ __forceinline__ void reduceThread(value_type val1, value_type val2)
  133. {
  134. mySum += __popc(val1 ^ val2);
  135. }
  136. __device__ __forceinline__ void reduceWarp(result_type* smem, uint tid)
  137. {
  138. warpReduce(smem, mySum, tid, plus<result_type>());
  139. }
  140. template <int THREAD_DIM> __device__ __forceinline__ void reduceBlock(result_type* smem, uint tid)
  141. {
  142. blockReduce<THREAD_DIM>(smem, mySum, tid, plus<result_type>());
  143. }
  144. __device__ __forceinline__ operator result_type() const
  145. {
  146. return mySum;
  147. }
  148. };
  149. //! @}
  150. }}
  151. #endif