3 "#line 1 \"libavfilter/opencl/nlmeans.cl\"\n" 5 " * This file is part of FFmpeg.\n" 7 " * FFmpeg is free software; you can redistribute it and/or\n" 8 " * modify it under the terms of the GNU Lesser General Public\n" 9 " * License as published by the Free Software Foundation; either\n" 10 " * version 2.1 of the License, or (at your option) any later version.\n" 12 " * FFmpeg is distributed in the hope that it will be useful,\n" 13 " * but WITHOUT ANY WARRANTY; without even the implied warranty of\n" 14 " * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU\n" 15 " * Lesser General Public License for more details.\n" 17 " * You should have received a copy of the GNU Lesser General Public\n" 18 " * License along with FFmpeg; if not, write to the Free Software\n" 19 " * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA\n" 22 "const sampler_t sampler = (CLK_NORMALIZED_COORDS_FALSE |\n" 23 " CLK_ADDRESS_CLAMP_TO_EDGE |\n" 24 " CLK_FILTER_NEAREST);\n" 26 "kernel void horiz_sum(__global uint4 *integral_img,\n" 27 " __read_only image2d_t src,\n" 34 " int y = get_global_id(0);\n" 35 " int work_size = get_global_size(0);\n" 37 " uint4 sum = (uint4)(0);\n" 39 " for (int i = 0; i < width; i++) {\n" 40 " float s1 = read_imagef(src, sampler, (int2)(i, y)).x;\n" 41 " s2.x = read_imagef(src, sampler, (int2)(i + dx.x, y + dy.x)).x;\n" 42 " s2.y = read_imagef(src, sampler, (int2)(i + dx.y, y + dy.y)).x;\n" 43 " s2.z = read_imagef(src, sampler, (int2)(i + dx.z, y + dy.z)).x;\n" 44 " s2.w = read_imagef(src, sampler, (int2)(i + dx.w, y + dy.w)).x;\n" 45 " sum += convert_uint4((s1 - s2) * (s1 - s2) * 255 * 255);\n" 46 " integral_img[y * width + i] = sum;\n" 50 "kernel void vert_sum(__global uint4 *integral_img,\n" 51 " __global int *overflow,\n" 55 " int x = get_global_id(0);\n" 57 " for (int i = 0; i < height; i++) {\n" 58 " if (any((uint4)UINT_MAX - integral_img[i * width + x] < sum))\n" 59 " atomic_inc(overflow);\n" 60 " integral_img[i * width + x] += sum;\n" 61 " sum = integral_img[i * width + x];\n" 65 "kernel void weight_accum(global float *sum, global float *weight,\n" 66 " global uint4 *integral_img, __read_only image2d_t src,\n" 67 " int width, int height, int p, float h,\n" 68 " int4 dx, int4 dy)\n" 70 " // w(x) = integral_img(x-p, y-p) +\n" 71 " // integral_img(x+p, y+p) -\n" 72 " // integral_img(x+p, y-p) -\n" 73 " // integral_img(x-p, y+p)\n" 74 " // total_sum[x] += w(x, y) * src(x + dx, y + dy)\n" 75 " // total_weight += w(x, y)\n" 77 " int x = get_global_id(0);\n" 78 " int y = get_global_id(1);\n" 79 " int4 xoff = x + dx;\n" 80 " int4 yoff = y + dy;\n" 81 " uint4 a = 0, b = 0, c = 0, d = 0;\n" 82 " uint4 src_pix = 0;\n" 84 " // out-of-bounding-box?\n" 85 " int oobb = (x - p) < 0 || (y - p) < 0 || (y + p) >= height || (x + p) >= width;\n" 87 " src_pix.x = (int)(255 * read_imagef(src, sampler, (int2)(xoff.x, yoff.x)).x);\n" 88 " src_pix.y = (int)(255 * read_imagef(src, sampler, (int2)(xoff.y, yoff.y)).x);\n" 89 " src_pix.z = (int)(255 * read_imagef(src, sampler, (int2)(xoff.z, yoff.z)).x);\n" 90 " src_pix.w = (int)(255 * read_imagef(src, sampler, (int2)(xoff.w, yoff.w)).x);\n" 92 " a = integral_img[(y - p) * width + x - p];\n" 93 " b = integral_img[(y + p) * width + x - p];\n" 94 " c = integral_img[(y - p) * width + x + p];\n" 95 " d = integral_img[(y + p) * width + x + p];\n" 98 " float4 patch_diff = convert_float4(d + a - c - b);\n" 99 " float4 w = native_exp(-patch_diff / (h * h));\n" 100 " float w_sum = w.x + w.y + w.z + w.w;\n" 101 " weight[y * width + x] += w_sum;\n" 102 " sum[y * width + x] += dot(w, convert_float4(src_pix));\n" 105 "kernel void average(__write_only image2d_t dst,\n" 106 " __read_only image2d_t src,\n" 107 " global float *sum, global float *weight) {\n" 108 " int x = get_global_id(0);\n" 109 " int y = get_global_id(1);\n" 110 " int2 dim = get_image_dim(dst);\n" 112 " float w = weight[y * dim.x + x];\n" 113 " float s = sum[y * dim.x + x];\n" 114 " float src_pix = read_imagef(src, sampler, (int2)(x, y)).x;\n" 115 " float r = (s + src_pix * 255) / (1.0f + w) / 255.0f;\n" 116 " if (x < dim.x && y < dim.y)\n" 117 " write_imagef(dst, (int2)(x, y), (float4)(r, 0.0f, 0.0f, 1.0f));\n" const char * ff_opencl_source_nlmeans