vf_crop.c 11.4 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
#include <stdio.h>

28
#include "avfilter.h"
29
#include "formats.h"
30
#include "internal.h"
31
#include "video.h"
32 33
#include "libavutil/eval.h"
#include "libavutil/avstring.h"
34
#include "libavutil/internal.h"
35
#include "libavutil/libm.h"
36
#include "libavutil/imgutils.h"
37
#include "libavutil/mathematics.h"
38
#include "libavutil/opt.h"
39

40
static const char *const var_names[] = {
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
    "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 {
57 58 59 60 61 62 63 64 65 66 67 68
    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_T,
    VAR_VARS_NB
69 70
};

71
typedef struct CropContext {
72
    const AVClass *class;
73 74 75 76 77
    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

78
    int max_step[4];    ///< max pixel step for each plane, expressed as a number of bytes
79
    int hsub, vsub;     ///< chroma subsampling
80
    char *x_expr, *y_expr, *ow_expr, *oh_expr;
81
    AVExpr *x_pexpr, *y_pexpr;  /* parsed expressions for x and y */
82
    double var_values[VAR_VARS_NB];
83 84 85 86
} CropContext;

static int query_formats(AVFilterContext *ctx)
{
87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110
    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
111 112
    };

113
    ff_set_common_formats(ctx, ff_make_format_list(pix_fmts));
114 115 116 117

    return 0;
}

118 119
static av_cold void uninit(AVFilterContext *ctx)
{
120
    CropContext *s = ctx->priv;
121

122 123 124 125
    av_expr_free(s->x_pexpr);
    s->x_pexpr = NULL;
    av_expr_free(s->y_pexpr);
    s->y_pexpr = NULL;
126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142
}

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;
}

143 144 145
static int config_input(AVFilterLink *link)
{
    AVFilterContext *ctx = link->dst;
146
    CropContext *s = ctx->priv;
147
    const AVPixFmtDescriptor *pix_desc = av_pix_fmt_desc_get(link->format);
148 149 150 151
    int ret;
    const char *expr;
    double res;

152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169
    s->var_values[VAR_E]     = M_E;
    s->var_values[VAR_PHI]   = M_PHI;
    s->var_values[VAR_PI]    = M_PI;
    s->var_values[VAR_IN_W]  = s->var_values[VAR_IW] = ctx->inputs[0]->w;
    s->var_values[VAR_IN_H]  = s->var_values[VAR_IH] = ctx->inputs[0]->h;
    s->var_values[VAR_X]     = NAN;
    s->var_values[VAR_Y]     = NAN;
    s->var_values[VAR_OUT_W] = s->var_values[VAR_OW] = NAN;
    s->var_values[VAR_OUT_H] = s->var_values[VAR_OH] = NAN;
    s->var_values[VAR_N]     = 0;
    s->var_values[VAR_T]     = NAN;

    av_image_fill_max_pixsteps(s->max_step, NULL, pix_desc);
    s->hsub = pix_desc->log2_chroma_w;
    s->vsub = pix_desc->log2_chroma_h;

    if ((ret = av_expr_parse_and_eval(&res, (expr = s->ow_expr),
                                      var_names, s->var_values,
170 171
                                      NULL, NULL, NULL, NULL, NULL, 0, ctx)) < 0)
        goto fail_expr;
172 173 174
    s->var_values[VAR_OUT_W] = s->var_values[VAR_OW] = res;
    if ((ret = av_expr_parse_and_eval(&res, (expr = s->oh_expr),
                                      var_names, s->var_values,
175 176
                                      NULL, NULL, NULL, NULL, NULL, 0, ctx)) < 0)
        goto fail_expr;
177
    s->var_values[VAR_OUT_H] = s->var_values[VAR_OH] = res;
178
    /* evaluate again ow as it may depend on oh */
179 180
    if ((ret = av_expr_parse_and_eval(&res, (expr = s->ow_expr),
                                      var_names, s->var_values,
181 182 183
                                      NULL, NULL, NULL, NULL, NULL, 0, ctx)) < 0)
        goto fail_expr;

184 185 186
    s->var_values[VAR_OUT_W] = s->var_values[VAR_OW] = res;
    if (normalize_double(&s->w, s->var_values[VAR_OUT_W]) < 0 ||
        normalize_double(&s->h, s->var_values[VAR_OUT_H]) < 0) {
187 188 189
        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",
190
               s->ow_expr, s->oh_expr);
191 192
        return AVERROR(EINVAL);
    }
193 194
    s->w &= ~((1 << s->hsub) - 1);
    s->h &= ~((1 << s->vsub) - 1);
195

196 197 198
    av_expr_free(s->x_pexpr);
    av_expr_free(s->y_pexpr);
    s->x_pexpr = s->y_pexpr = NULL;
199
    if ((ret = av_expr_parse(&s->x_pexpr, s->x_expr, var_names,
200
                             NULL, NULL, NULL, NULL, 0, ctx)) < 0 ||
201
        (ret = av_expr_parse(&s->y_pexpr, s->y_expr, var_names,
202 203
                             NULL, NULL, NULL, NULL, 0, ctx)) < 0)
        return AVERROR(EINVAL);
204

205
    av_log(ctx, AV_LOG_VERBOSE, "w:%d h:%d -> w:%d h:%d\n",
206
           link->w, link->h, s->w, s->h);
207

208 209
    if (s->w <= 0 || s->h <= 0 ||
        s->w > link->w || s->h > link->h) {
210
        av_log(ctx, AV_LOG_ERROR,
211
               "Invalid too big or non positive size for width '%d' or height '%d'\n",
212
               s->w, s->h);
213
        return AVERROR(EINVAL);
214 215
    }

216
    /* set default, required in the case the first computed value for x/y is NAN */
217 218 219 220
    s->x = (link->w - s->w) / 2;
    s->y = (link->h - s->h) / 2;
    s->x &= ~((1 << s->hsub) - 1);
    s->y &= ~((1 << s->vsub) - 1);
221
    return 0;
222 223 224 225

fail_expr:
    av_log(NULL, AV_LOG_ERROR, "Error when evaluating the expression '%s'\n", expr);
    return ret;
226 227 228 229
}

static int config_output(AVFilterLink *link)
{
230
    CropContext *s = link->src->priv;
231

232 233
    link->w = s->w;
    link->h = s->h;
234 235 236 237

    return 0;
}

238
static int filter_frame(AVFilterLink *link, AVFrame *frame)
239
{
240
    AVFilterContext *ctx = link->dst;
241
    CropContext *s = ctx->priv;
242
    const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(link->format);
243 244
    int i;

245 246
    frame->width  = s->w;
    frame->height = s->h;
247

248
    s->var_values[VAR_T] = frame->pts == AV_NOPTS_VALUE ?
249
        NAN : frame->pts * av_q2d(link->time_base);
250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266
    s->var_values[VAR_X] = av_expr_eval(s->x_pexpr, s->var_values, NULL);
    s->var_values[VAR_Y] = av_expr_eval(s->y_pexpr, s->var_values, NULL);
    s->var_values[VAR_X] = av_expr_eval(s->x_pexpr, s->var_values, NULL);

    normalize_double(&s->x, s->var_values[VAR_X]);
    normalize_double(&s->y, s->var_values[VAR_Y]);

    if (s->x < 0)
        s->x = 0;
    if (s->y < 0)
        s->y = 0;
    if ((unsigned)s->x + (unsigned)s->w > link->w)
        s->x = link->w - s->w;
    if ((unsigned)s->y + (unsigned)s->h > link->h)
        s->y = link->h - s->h;
    s->x &= ~((1 << s->hsub) - 1);
    s->y &= ~((1 << s->vsub) - 1);
267

268
    av_dlog(ctx, "n:%d t:%f x:%d y:%d x+w:%d y+h:%d\n",
269 270
            (int)s->var_values[VAR_N], s->var_values[VAR_T], s->x,
            s->y, s->x+s->w, s->y+s->h);
271

272 273
    frame->data[0] += s->y * frame->linesize[0];
    frame->data[0] += s->x * s->max_step[0];
274

275
    if (!(desc->flags & AV_PIX_FMT_FLAG_PAL || desc->flags & AV_PIX_FMT_FLAG_PSEUDOPAL)) {
276
        for (i = 1; i < 3; i ++) {
277
            if (frame->data[i]) {
278 279
                frame->data[i] += (s->y >> s->vsub) * frame->linesize[i];
                frame->data[i] += (s->x * s->max_step[i]) >> s->hsub;
280 281 282 283 284
            }
        }
    }

    /* alpha plane */
285
    if (frame->data[3]) {
286 287
        frame->data[3] += s->y * frame->linesize[3];
        frame->data[3] += s->x * s->max_step[3];
288 289
    }

290
    s->var_values[VAR_N] += 1.0;
291 292

    return ff_filter_frame(link->dst->outputs[0], frame);
293 294
}

295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313
#define OFFSET(x) offsetof(CropContext, x)
#define FLAGS AV_OPT_FLAG_VIDEO_PARAM
static const AVOption options[] = {
    { "out_w", "Output video width",  OFFSET(ow_expr), AV_OPT_TYPE_STRING, { .str = "iw" },                 .flags = FLAGS },
    { "out_h", "Output video height", OFFSET(oh_expr), AV_OPT_TYPE_STRING, { .str = "ih" },                 .flags = FLAGS },
    { "x",     "Horizontal position in the input video of the left edge of the cropped output video",
                                      OFFSET(x_expr),  AV_OPT_TYPE_STRING, { .str = "(in_w - out_w) / 2" }, .flags = FLAGS },
    { "y",     "Vertical position in the input video of the top edge of the cropped output video",
                                      OFFSET(y_expr),  AV_OPT_TYPE_STRING, { .str = "(in_h - out_h) / 2" }, .flags = FLAGS },
    { NULL },
};

static const AVClass crop_class = {
    .class_name = "crop",
    .item_name  = av_default_item_name,
    .option     = options,
    .version    = LIBAVUTIL_VERSION_INT,
};

314 315 316 317
static const AVFilterPad avfilter_vf_crop_inputs[] = {
    {
        .name             = "default",
        .type             = AVMEDIA_TYPE_VIDEO,
318
        .filter_frame     = filter_frame,
319 320 321 322 323 324 325 326 327 328 329 330 331 332 333
        .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 }
};

334
AVFilter ff_vf_crop = {
335
    .name      = "crop",
336
    .description = NULL_IF_CONFIG_SMALL("Crop the input video to width:height:x:y."),
337

338
    .priv_size = sizeof(CropContext),
339
    .priv_class = &crop_class,
340 341

    .query_formats = query_formats,
342
    .uninit        = uninit,
343

344 345
    .inputs    = avfilter_vf_crop_inputs,
    .outputs   = avfilter_vf_crop_outputs,
346
};