On Thu, Apr 11, 2013 at 03:44:48PM +0200, Nicolas Bertrand wrote:
> For testing purpose. Used to test XYZ JPEG2000 output
> ---
> libavfilter/Makefile | 1 +
> libavfilter/allfilters.c | 1 +
> libavfilter/vf_xyz2rgb.c | 201
> ++++++++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 203 insertions(+)
> create mode 100644 libavfilter/vf_xyz2rgb.c
>
> diff --git a/libavfilter/Makefile b/libavfilter/Makefile
> index 4a3331a..816c640 100644
> --- a/libavfilter/Makefile
> +++ b/libavfilter/Makefile
> @@ -77,6 +77,7 @@ OBJS-$(CONFIG_SPLIT_FILTER) += split.o
> OBJS-$(CONFIG_TRANSPOSE_FILTER) += vf_transpose.o
> OBJS-$(CONFIG_UNSHARP_FILTER) += vf_unsharp.o
> OBJS-$(CONFIG_VFLIP_FILTER) += vf_vflip.o
> +OBJS-$(CONFIG_XYZ2RGB_FILTER) += vf_xyz2rgb.o
> OBJS-$(CONFIG_YADIF_FILTER) += vf_yadif.o
>
> OBJS-$(CONFIG_COLOR_FILTER) += vsrc_color.o
> diff --git a/libavfilter/allfilters.c b/libavfilter/allfilters.c
> index f3ce91c..d1021b1 100644
> --- a/libavfilter/allfilters.c
> +++ b/libavfilter/allfilters.c
> @@ -97,6 +97,7 @@ void avfilter_register_all(void)
> REGISTER_FILTER(TRANSPOSE, transpose, vf);
> REGISTER_FILTER(UNSHARP, unsharp, vf);
> REGISTER_FILTER(VFLIP, vflip, vf);
> + REGISTER_FILTER(XYZ2RGB, xyz2rgb, vf);
> REGISTER_FILTER(YADIF, yadif, vf);
>
> REGISTER_FILTER(COLOR, color, vsrc);
> diff --git a/libavfilter/vf_xyz2rgb.c b/libavfilter/vf_xyz2rgb.c
> new file mode 100644
> index 0000000..866daa0
> --- /dev/null
> +++ b/libavfilter/vf_xyz2rgb.c
> @@ -0,0 +1,201 @@
> +/*
> + * XYZ to RGB filter
> + * Copyright (c) 2012 Matthias Buercher
> + * Copyright (c) 2013 Nicolas Bertrand <[email protected]>
> + *
> + * This file is part of Libav.
> + *
> + * Libav 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.
> + *
> + * Libav 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 Libav; if not, write to the Free Software
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301
> USA
> + */
> +
> +/**
> + * @file
> + * xyz2rgb filter
> + * Converts from XYZ to RGB space
> + * Useful to convert jpeg2000 files from MXF containers in DCP
> + * The filter has no parameters
> + */
> +
> +#include <string.h>
> +
> +#include "libavutil/pixdesc.h"
> +#include "libavutil/common.h"
> +#include "libavutil/internal.h"
> +#include "libavutil/intreadwrite.h"
> +#include "libavutil/imgutils.h"
> +#include "avfilter.h"
> +#include "formats.h"
> +#include "internal.h"
> +#include "video.h"
> +
> +/* pre defined color-spaces gamma */
> +#define XYZ_GAMMA (2.6f)
> +#define RGB_GAMMA (2.2f)
> +
> +typedef struct {
> + int xyzgamma[4096];
> + int rgbgamma[4096];
> + int (* matrix)[3];
> +} XYZ2RGBContext;
> +
> +/*
> + * XYZ to RGB conversion matrix
> + * R | 3.2404542 -1.5371385 -0.4985314 | X
> + * G = 4095 * | -0.9692660 1.8760108 0.0415560 | Y
> + * B | 0.0556434 -0.2040259 1.0572252 | Z
> + *
> + * The matrix is stored in int values
> + */
> +static int xyz2rgb_matrix[3][3] = {
> + {13270, -6295, -2041},
> + {-3969, 7682, 170},
> + { 228, -835, 4329} };
> +
> +static int query_formats(AVFilterContext * ctx)
> +{
> + AVFilterFormats *formats;
> + int ret = -1 ;
> +
> + if (ctx->inputs[0]) {
> + formats = NULL;
> + ret = ff_add_format(&formats, AV_PIX_FMT_XYZ12);
> + ff_formats_ref(formats, &ctx->inputs[0]->out_formats);
Add both XYZ12_LE and _BE - otherwise it can't be used on big-endian platforms
(where we test it as well).
> + }
> + if (ctx->outputs[0]) {
> + formats = NULL;
> + ret = ff_add_format(&formats, AV_PIX_FMT_RGB48);
> + ff_formats_ref(formats, &ctx->outputs[0]->in_formats);
> + }
> + return ret;
> +}
> +
> +/**
> + * The gamma values are precalculated in an array
> + * XYZ uses projector gamma 2.6
> + * sRGB uses gamma 2.2
> + * The gamma function is the inverse power function, calculated in [0..1]
> and scaled to 12-bit fixed point
> + *
> + * The matrix multipliers are precalculated and scaled to 12bit bitdepth
> (4096 values)
> + * For documentation see
> + * http://www.brucelindbloom.com/index.html?Eqn_RGB_XYZ_Matrix.html
> + * http://en.wikipedia.org/wiki/SRGB
> +*/
> +
> +static int config_props(AVFilterLink * inlink)
> +{
> + XYZ2RGBContext *settings = inlink->dst->priv;
> + int i;
> + double xyzgamma = XYZ_GAMMA;
> + double rgbgamma = 1.0 / RGB_GAMMA;
> +
> + /* set gamma vectors */
> + for (i = 0; i < 4096; i++) {
> + settings->xyzgamma[i] =
> + lrintf(pow(i / 4095.0, xyzgamma) * 4095.0);
> + settings->rgbgamma[i] =
> + lrintf(pow(i / 4095.0, rgbgamma) * 4095.0);
> + }
> +
> + settings->matrix = xyz2rgb_matrix;
> + return 0;
> +}
> +
> +static int filter_frame(AVFilterLink * inlink, AVFrame * in)
> +{
> + AVFilterLink *outlink = inlink->dst->outputs[0];
> + XYZ2RGBContext *settings = inlink->dst->priv;
> +
> + AVFrame *out;
> + uint8_t *inrow, *outrow;
> + int i, j;
> + int r, g, b, x, y, z;
> +
> + out = ff_get_video_buffer(outlink, outlink->w, outlink->h);
> + if (!out) {
> + av_frame_free(&in);
> + return AVERROR(ENOMEM);
> + }
> +
> + out->pts = in->pts;
> + inrow = in->data[0];
> + outrow = out->data[0];
> +
> + /*
> + * In case of XYZ (12-bit depth) format input
> + * conversion to RGB24 (8-bit depth) --> loss of precision
> + * FIXME: test/implement output RGB48 for display
> + */
> + if (inlink->format == AV_PIX_FMT_XYZ12) {
looks like a useless check - what other format can you get here?
> + for (i = 0; i < inlink->h; i++) {
> + for (j = 0; j < inlink->w * 6; j += 6) {
> + // read little endian and scale from 16bit to 12bit
> + x = AV_RL16(inrow + j) >> 4;
> + y = AV_RL16(inrow + j + 2) >> 4;
> + z = AV_RL16(inrow + j + 4) >> 4;
> +
> + // convert from XYZ to XYZlinear
> + x = settings->xyzgamma[x];
> + y = settings->xyzgamma[y];
> + z = settings->xyzgamma[z];
> +
> + // convert from XYZlinear to sRGBlinear
> + r = settings->matrix[0][0] * x +
> + settings->matrix[0][1] * y +
> + settings->matrix[0][2] * z >> 12;
> + g = settings->matrix[1][0] * x +
> + settings->matrix[1][1] * y +
> + settings->matrix[1][2] * z >> 12;
> + b = settings->matrix[2][0] * x +
> + settings->matrix[1][2] * y +
> + settings->matrix[2][2] * z >> 12;
> +
> + // limit values to 12-bit depth
> + r = av_clip_c(r,0,4095);
> + g = av_clip_c(g,0,4095);
> + b = av_clip_c(b,0,4095);
> +
> + // convert from sRGBlinear to RGB and scale from 12bit to
> 16bit
> + r = settings->rgbgamma[r] << 4;
> + g = settings->rgbgamma[g] << 4;
> + b = settings->rgbgamma[b] << 4;
> +
> + // write little endian
> + AV_WL16(outrow + j, r);
> + AV_WL16(outrow + j + 2, g);
> + AV_WL16(outrow + j + 4, b);
> + }
> + inrow += in->linesize[0];
> + outrow += out->linesize[0];
Diego will punish you for wrong indentation and lack of vertical alignment
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
