vf_crop.c 11.2 KB
Newer Older
1
/*
2
 * Copyright (c) 2007 Bobby Bingham
3
 *
4
 * This file is part of Libav.
5
 *
6
 * Libav is free software; you can redistribute it and/or
7 8 9 10
 * 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.
 *
11
 * Libav is distributed in the hope that it will be useful,
12 13 14 15 16
 * 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
17
 * License along with Libav; if not, write to the Free Software
18 19 20 21
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
22
 * @file
23 24 25
 * video crop filter
 */

26 27
/* #define DEBUG */

28 29
#include <stdio.h>

30
#include "avfilter.h"
31
#include "formats.h"
32
#include "internal.h"
33
#include "video.h"
34 35
#include "libavutil/eval.h"
#include "libavutil/avstring.h"
36
#include "libavutil/internal.h"
37
#include "libavutil/libm.h"
38
#include "libavutil/imgutils.h"
39
#include "libavutil/mathematics.h"
40

41
static const char *const var_names[] = {
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
    "E",
    "PHI",
    "PI",
    "in_w", "iw",   ///< width  of the input video
    "in_h", "ih",   ///< height of the input video
    "out_w", "ow",  ///< width  of the cropped video
    "out_h", "oh",  ///< height of the cropped video
    "x",
    "y",
    "n",            ///< number of frame
    "pos",          ///< position in the file
    "t",            ///< timestamp expressed in seconds
    NULL
};

enum var_name {
58 59 60 61 62 63 64 65 66 67 68 69 70
    VAR_E,
    VAR_PHI,
    VAR_PI,
    VAR_IN_W,  VAR_IW,
    VAR_IN_H,  VAR_IH,
    VAR_OUT_W, VAR_OW,
    VAR_OUT_H, VAR_OH,
    VAR_X,
    VAR_Y,
    VAR_N,
    VAR_POS,
    VAR_T,
    VAR_VARS_NB
71 72
};

73 74 75 76 77 78
typedef struct {
    int  x;             ///< x offset of the non-cropped area with respect to the input area
    int  y;             ///< y offset of the non-cropped area with respect to the input area
    int  w;             ///< width of the cropped area
    int  h;             ///< height of the cropped area

79
    int max_step[4];    ///< max pixel step for each plane, expressed as a number of bytes
80
    int hsub, vsub;     ///< chroma subsampling
81 82
    char x_expr[256], y_expr[256], ow_expr[256], oh_expr[256];
    AVExpr *x_pexpr, *y_pexpr;  /* parsed expressions for x and y */
83
    double var_values[VAR_VARS_NB];
84 85 86 87
} CropContext;

static int query_formats(AVFilterContext *ctx)
{
88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
    static const enum AVPixelFormat pix_fmts[] = {
        AV_PIX_FMT_RGB48BE,      AV_PIX_FMT_RGB48LE,
        AV_PIX_FMT_BGR48BE,      AV_PIX_FMT_BGR48LE,
        AV_PIX_FMT_ARGB,         AV_PIX_FMT_RGBA,
        AV_PIX_FMT_ABGR,         AV_PIX_FMT_BGRA,
        AV_PIX_FMT_RGB24,        AV_PIX_FMT_BGR24,
        AV_PIX_FMT_RGB565BE,     AV_PIX_FMT_RGB565LE,
        AV_PIX_FMT_RGB555BE,     AV_PIX_FMT_RGB555LE,
        AV_PIX_FMT_BGR565BE,     AV_PIX_FMT_BGR565LE,
        AV_PIX_FMT_BGR555BE,     AV_PIX_FMT_BGR555LE,
        AV_PIX_FMT_GRAY16BE,     AV_PIX_FMT_GRAY16LE,
        AV_PIX_FMT_YUV420P16LE,  AV_PIX_FMT_YUV420P16BE,
        AV_PIX_FMT_YUV422P16LE,  AV_PIX_FMT_YUV422P16BE,
        AV_PIX_FMT_YUV444P16LE,  AV_PIX_FMT_YUV444P16BE,
        AV_PIX_FMT_YUV444P,      AV_PIX_FMT_YUV422P,
        AV_PIX_FMT_YUV420P,      AV_PIX_FMT_YUV411P,
        AV_PIX_FMT_YUV410P,      AV_PIX_FMT_YUV440P,
        AV_PIX_FMT_YUVJ444P,     AV_PIX_FMT_YUVJ422P,
        AV_PIX_FMT_YUVJ420P,     AV_PIX_FMT_YUVJ440P,
        AV_PIX_FMT_YUVA420P,
        AV_PIX_FMT_RGB8,         AV_PIX_FMT_BGR8,
        AV_PIX_FMT_RGB4_BYTE,    AV_PIX_FMT_BGR4_BYTE,
        AV_PIX_FMT_PAL8,         AV_PIX_FMT_GRAY8,
        AV_PIX_FMT_NONE
112 113
    };

114
    ff_set_common_formats(ctx, ff_make_format_list(pix_fmts));
115 116 117 118

    return 0;
}

119
static av_cold int init(AVFilterContext *ctx, const char *args)
120 121 122
{
    CropContext *crop = ctx->priv;

123 124 125 126 127
    av_strlcpy(crop->ow_expr, "iw", sizeof(crop->ow_expr));
    av_strlcpy(crop->oh_expr, "ih", sizeof(crop->oh_expr));
    av_strlcpy(crop->x_expr, "(in_w-out_w)/2", sizeof(crop->x_expr));
    av_strlcpy(crop->y_expr, "(in_h-out_h)/2", sizeof(crop->y_expr));

128
    if (args)
129
        sscanf(args, "%255[^:]:%255[^:]:%255[^:]:%255[^:]", crop->ow_expr, crop->oh_expr, crop->x_expr, crop->y_expr);
130 131 132 133

    return 0;
}

134 135 136 137
static av_cold void uninit(AVFilterContext *ctx)
{
    CropContext *crop = ctx->priv;

138 139
    av_expr_free(crop->x_pexpr); crop->x_pexpr = NULL;
    av_expr_free(crop->y_pexpr); crop->y_pexpr = NULL;
140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156
}

static inline int normalize_double(int *n, double d)
{
    int ret = 0;

    if (isnan(d)) {
        ret = AVERROR(EINVAL);
    } else if (d > INT_MAX || d < INT_MIN) {
        *n = d > INT_MAX ? INT_MAX : INT_MIN;
        ret = AVERROR(EINVAL);
    } else
        *n = round(d);

    return ret;
}

157 158 159 160
static int config_input(AVFilterLink *link)
{
    AVFilterContext *ctx = link->dst;
    CropContext *crop = ctx->priv;
161
    const AVPixFmtDescriptor *pix_desc = av_pix_fmt_desc_get(link->format);
162 163 164 165
    int ret;
    const char *expr;
    double res;

166 167 168 169 170 171 172 173 174 175 176 177
    crop->var_values[VAR_E]     = M_E;
    crop->var_values[VAR_PHI]   = M_PHI;
    crop->var_values[VAR_PI]    = M_PI;
    crop->var_values[VAR_IN_W]  = crop->var_values[VAR_IW] = ctx->inputs[0]->w;
    crop->var_values[VAR_IN_H]  = crop->var_values[VAR_IH] = ctx->inputs[0]->h;
    crop->var_values[VAR_X]     = NAN;
    crop->var_values[VAR_Y]     = NAN;
    crop->var_values[VAR_OUT_W] = crop->var_values[VAR_OW] = NAN;
    crop->var_values[VAR_OUT_H] = crop->var_values[VAR_OH] = NAN;
    crop->var_values[VAR_N]     = 0;
    crop->var_values[VAR_T]     = NAN;
    crop->var_values[VAR_POS]   = NAN;
178

179
    av_image_fill_max_pixsteps(crop->max_step, NULL, pix_desc);
180 181
    crop->hsub = pix_desc->log2_chroma_w;
    crop->vsub = pix_desc->log2_chroma_h;
182

183
    if ((ret = av_expr_parse_and_eval(&res, (expr = crop->ow_expr),
184 185
                                      var_names, crop->var_values,
                                      NULL, NULL, NULL, NULL, NULL, 0, ctx)) < 0) goto fail_expr;
186
    crop->var_values[VAR_OUT_W] = crop->var_values[VAR_OW] = res;
187
    if ((ret = av_expr_parse_and_eval(&res, (expr = crop->oh_expr),
188 189
                                      var_names, crop->var_values,
                                      NULL, NULL, NULL, NULL, NULL, 0, ctx)) < 0) goto fail_expr;
190
    crop->var_values[VAR_OUT_H] = crop->var_values[VAR_OH] = res;
191
    /* evaluate again ow as it may depend on oh */
192
    if ((ret = av_expr_parse_and_eval(&res, (expr = crop->ow_expr),
193 194
                                      var_names, crop->var_values,
                                      NULL, NULL, NULL, NULL, NULL, 0, ctx)) < 0) goto fail_expr;
195 196 197
    crop->var_values[VAR_OUT_W] = crop->var_values[VAR_OW] = res;
    if (normalize_double(&crop->w, crop->var_values[VAR_OUT_W]) < 0 ||
        normalize_double(&crop->h, crop->var_values[VAR_OUT_H]) < 0) {
198 199 200 201 202 203 204 205
        av_log(ctx, AV_LOG_ERROR,
               "Too big value or invalid expression for out_w/ow or out_h/oh. "
               "Maybe the expression for out_w:'%s' or for out_h:'%s' is self-referencing.\n",
               crop->ow_expr, crop->oh_expr);
        return AVERROR(EINVAL);
    }
    crop->w &= ~((1 << crop->hsub) - 1);
    crop->h &= ~((1 << crop->vsub) - 1);
206

207
    if ((ret = av_expr_parse(&crop->x_pexpr, crop->x_expr, var_names,
208
                             NULL, NULL, NULL, NULL, 0, ctx)) < 0 ||
209
        (ret = av_expr_parse(&crop->y_pexpr, crop->y_expr, var_names,
210 211
                             NULL, NULL, NULL, NULL, 0, ctx)) < 0)
        return AVERROR(EINVAL);
212

213
    av_log(ctx, AV_LOG_VERBOSE, "w:%d h:%d -> w:%d h:%d\n",
214
           link->w, link->h, crop->w, crop->h);
215

216 217
    if (crop->w <= 0 || crop->h <= 0 ||
        crop->w > link->w || crop->h > link->h) {
218
        av_log(ctx, AV_LOG_ERROR,
219 220
               "Invalid too big or non positive size for width '%d' or height '%d'\n",
               crop->w, crop->h);
221
        return AVERROR(EINVAL);
222 223
    }

224 225 226 227 228
    /* set default, required in the case the first computed value for x/y is NAN */
    crop->x = (link->w - crop->w) / 2;
    crop->y = (link->h - crop->h) / 2;
    crop->x &= ~((1 << crop->hsub) - 1);
    crop->y &= ~((1 << crop->vsub) - 1);
229
    return 0;
230 231 232 233

fail_expr:
    av_log(NULL, AV_LOG_ERROR, "Error when evaluating the expression '%s'\n", expr);
    return ret;
234 235 236 237 238 239 240 241 242 243 244 245
}

static int config_output(AVFilterLink *link)
{
    CropContext *crop = link->src->priv;

    link->w = crop->w;
    link->h = crop->h;

    return 0;
}

246
static int filter_frame(AVFilterLink *link, AVFilterBufferRef *frame)
247
{
248 249
    AVFilterContext *ctx = link->dst;
    CropContext *crop = ctx->priv;
250
    const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(link->format);
251 252
    int i;

253 254
    frame->video->w = crop->w;
    frame->video->h = crop->h;
255

256 257 258
    crop->var_values[VAR_T] = frame->pts == AV_NOPTS_VALUE ?
        NAN : frame->pts * av_q2d(link->time_base);
    crop->var_values[VAR_POS] = frame->pos == -1 ? NAN : frame->pos;
259 260 261
    crop->var_values[VAR_X] = av_expr_eval(crop->x_pexpr, crop->var_values, NULL);
    crop->var_values[VAR_Y] = av_expr_eval(crop->y_pexpr, crop->var_values, NULL);
    crop->var_values[VAR_X] = av_expr_eval(crop->x_pexpr, crop->var_values, NULL);
262

263 264
    normalize_double(&crop->x, crop->var_values[VAR_X]);
    normalize_double(&crop->y, crop->var_values[VAR_Y]);
265 266 267 268 269 270 271 272

    if (crop->x < 0) crop->x = 0;
    if (crop->y < 0) crop->y = 0;
    if ((unsigned)crop->x + (unsigned)crop->w > link->w) crop->x = link->w - crop->w;
    if ((unsigned)crop->y + (unsigned)crop->h > link->h) crop->y = link->h - crop->h;
    crop->x &= ~((1 << crop->hsub) - 1);
    crop->y &= ~((1 << crop->vsub) - 1);

273 274 275
    av_dlog(ctx, "n:%d t:%f x:%d y:%d x+w:%d y+h:%d\n",
            (int)crop->var_values[VAR_N], crop->var_values[VAR_T], crop->x,
            crop->y, crop->x+crop->w, crop->y+crop->h);
276

277 278
    frame->data[0] += crop->y * frame->linesize[0];
    frame->data[0] += crop->x * crop->max_step[0];
279

280
    if (!(desc->flags & PIX_FMT_PAL || desc->flags & PIX_FMT_PSEUDOPAL)) {
281
        for (i = 1; i < 3; i ++) {
282 283 284
            if (frame->data[i]) {
                frame->data[i] += (crop->y >> crop->vsub) * frame->linesize[i];
                frame->data[i] += (crop->x * crop->max_step[i]) >> crop->hsub;
285 286 287 288 289
            }
        }
    }

    /* alpha plane */
290 291 292
    if (frame->data[3]) {
        frame->data[3] += crop->y * frame->linesize[3];
        frame->data[3] += crop->x * crop->max_step[3];
293 294
    }

295
    crop->var_values[VAR_N] += 1.0;
296 297

    return ff_filter_frame(link->dst->outputs[0], frame);
298 299
}

300 301 302 303
static const AVFilterPad avfilter_vf_crop_inputs[] = {
    {
        .name             = "default",
        .type             = AVMEDIA_TYPE_VIDEO,
304
        .filter_frame     = filter_frame,
305 306 307 308 309 310 311 312 313 314 315 316 317 318 319
        .get_video_buffer = ff_null_get_video_buffer,
        .config_props     = config_input,
    },
    { NULL }
};

static const AVFilterPad avfilter_vf_crop_outputs[] = {
    {
        .name         = "default",
        .type         = AVMEDIA_TYPE_VIDEO,
        .config_props = config_output,
    },
    { NULL }
};

320 321
AVFilter avfilter_vf_crop = {
    .name      = "crop",
322
    .description = NULL_IF_CONFIG_SMALL("Crop the input video to width:height:x:y."),
323

324 325 326 327
    .priv_size = sizeof(CropContext),

    .query_formats = query_formats,
    .init          = init,
328
    .uninit        = uninit,
329

330 331
    .inputs    = avfilter_vf_crop_inputs,
    .outputs   = avfilter_vf_crop_outputs,
332
};