aboutsummaryrefslogtreecommitdiffstats
path: root/libavfilter/opencl/convolution.cl
blob: 815c7792725923bf7eda148da7aa0dd2bb26791f (plain) (blame)
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
/*
 * Copyright (c) 2018 Danil Iashchenko
 *
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * FFmpeg is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

__kernel void convolution_global(__write_only image2d_t dst,
                                 __read_only  image2d_t src,
                                 int coef_matrix_dim,
                                 __constant float *coef_matrix,
                                 float div,
                                 float bias)
{
    const sampler_t sampler = (CLK_NORMALIZED_COORDS_FALSE |
                               CLK_ADDRESS_CLAMP_TO_EDGE   |
                               CLK_FILTER_NEAREST);

    const int half_matrix_dim = (coef_matrix_dim / 2);
    int2 loc = (int2)(get_global_id(0), get_global_id(1));
    float4 convPix = (float4)(0.0f, 0.0f, 0.0f, 0.0f);

    for (int conv_i = -half_matrix_dim; conv_i <= half_matrix_dim; conv_i++) {
        for (int conv_j = -half_matrix_dim; conv_j <= half_matrix_dim; conv_j++) {
            float4 px = read_imagef(src, sampler, loc + (int2)(conv_j, conv_i));
            convPix += px * coef_matrix[(conv_i + half_matrix_dim) * coef_matrix_dim +
                                        (conv_j + half_matrix_dim)];
        }
     }
     float4 dstPix = convPix * div + bias;
     write_imagef(dst, loc, dstPix);
}


__kernel void sobel_global(__write_only image2d_t dst,
                           __read_only  image2d_t src,
                             float div,
                             float bias)
{
    const sampler_t sampler = (CLK_NORMALIZED_COORDS_FALSE |
                               CLK_ADDRESS_CLAMP_TO_EDGE   |
                               CLK_FILTER_NEAREST);

    int2 loc = (int2)(get_global_id(0), get_global_id(1));

    float4 sum1 = read_imagef(src, sampler, loc + (int2)(-1,-1)) * -1 +
                  read_imagef(src, sampler, loc + (int2)( 0,-1)) * -2 +
                  read_imagef(src, sampler, loc + (int2)( 1,-1)) * -1 +
                  read_imagef(src, sampler, loc + (int2)(-1, 1)) *  1 +
                  read_imagef(src, sampler, loc + (int2)( 0, 1)) *  2 +
                  read_imagef(src, sampler, loc + (int2)( 1, 1)) *  1;

    float4 sum2 = read_imagef(src, sampler, loc + (int2)(-1,-1)) * -1 +
                  read_imagef(src, sampler, loc + (int2)(-1, 0)) * -2 +
                  read_imagef(src, sampler, loc + (int2)(-1, 1)) * -1 +
                  read_imagef(src, sampler, loc + (int2)( 1,-1)) *  1 +
                  read_imagef(src, sampler, loc + (int2)( 1, 0)) *  2 +
                  read_imagef(src, sampler, loc + (int2)( 1, 1)) *  1;

    float4 dstPix = hypot(sum1, sum2) * div + bias;
    write_imagef(dst, loc, dstPix);
}

__kernel void prewitt_global(__write_only image2d_t dst,
                             __read_only  image2d_t src,
                             float div,
                             float bias)
{
    const sampler_t sampler = (CLK_NORMALIZED_COORDS_FALSE |
                               CLK_ADDRESS_CLAMP_TO_EDGE   |
                               CLK_FILTER_NEAREST);

    int2 loc = (int2)(get_global_id(0), get_global_id(1));

    float4 sum1 = read_imagef(src, sampler, loc + (int2)(-1,-1)) *  1 +
                  read_imagef(src, sampler, loc + (int2)( 0,-1)) *  1 +
                  read_imagef(src, sampler, loc + (int2)( 1,-1)) *  1 +
                  read_imagef(src, sampler, loc + (int2)(-1, 1)) * -1 +
                  read_imagef(src, sampler, loc + (int2)( 0, 1)) * -1 +
                  read_imagef(src, sampler, loc + (int2)( 1, 1)) * -1;

    float4 sum2 = read_imagef(src, sampler, loc + (int2)(-1,-1)) *  1 +
                  read_imagef(src, sampler, loc + (int2)(-1, 0)) *  1 +
                  read_imagef(src, sampler, loc + (int2)(-1, 1)) *  1 +
                  read_imagef(src, sampler, loc + (int2)( 1,-1)) * -1 +
                  read_imagef(src, sampler, loc + (int2)( 1, 0)) * -1 +
                  read_imagef(src, sampler, loc + (int2)( 1, 1)) * -1;

    float4 dstPix = hypot(sum1, sum2) * div + bias;
    write_imagef(dst, loc, dstPix);
}

__kernel void roberts_global(__write_only image2d_t dst,
                             __read_only  image2d_t src,
                             float div,
                             float bias)
{
    const sampler_t sampler = (CLK_NORMALIZED_COORDS_FALSE |
                               CLK_ADDRESS_CLAMP_TO_EDGE   |
                               CLK_FILTER_NEAREST);

    int2 loc = (int2)(get_global_id(0), get_global_id(1));

    float4 sum1 = read_imagef(src, sampler, loc + (int2)(-1,-1)) *  1 +
                  read_imagef(src, sampler, loc + (int2)( 0,-1)) * -1;


    float4 sum2 = read_imagef(src, sampler, loc + (int2)(-1, 0)) * -1 +
                  read_imagef(src, sampler, loc + (int2)( 0, 0)) *  1;


    float4 dstPix = hypot(sum1, sum2) * div + bias;
    write_imagef(dst, loc, dstPix);
}