FFmpeg/libavfilter/vf_framerate.c

/*
 * Copyright (C) 2012 Mark Himsley
 *
 * get_scene_score() Copyright (c) 2011 Stefano Sabatini
 * taken from libavfilter/vf_select.c
 *
 * 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
 */

/**
 * @file
 * filter for upsampling or downsampling a progressive source
 */

#define DEBUG

#include "libavutil/avassert.h"
#include "libavutil/imgutils.h"
#include "libavutil/internal.h"
#include "libavutil/opt.h"
#include "libavutil/pixdesc.h"
#include "libavutil/pixelutils.h"

#include "avfilter.h"
#include "internal.h"
#include "video.h"

#define N_SRCE 3

typedef struct FrameRateContext {
    const AVClass *class;
    // parameters
    AVRational dest_frame_rate;         ///< output frames per second
    int flags;                          ///< flags affecting frame rate conversion algorithm
    double scene_score;                 ///< score that denotes a scene change has happened
    int interp_start;                   ///< start of range to apply linear interpolation
    int interp_end;                     ///< end of range to apply linear interpolation

    int line_size[4];                   ///< bytes of pixel data per line for each plane
    int vsub;

    int frst, next, prev, crnt, last;
    int pending_srce_frames;            ///< how many input frames are still waiting to be processed
    int flush;                          ///< are we flushing final frames
    int pending_end_frame;              ///< flag indicating we are waiting to call filter_frame()

    AVRational srce_time_base;          ///< timebase of source

    AVRational dest_time_base;          ///< timebase of destination
    int32_t dest_frame_num;
    int64_t last_dest_frame_pts;        ///< pts of the last frame output
    int64_t average_srce_pts_dest_delta;///< average input pts delta converted from input rate to output rate
    int64_t average_dest_pts_delta;     ///< calculated average output pts delta

    av_pixelutils_sad_fn sad;           ///< Sum of the absolute difference function (scene detect only)
    double prev_mafd;                   ///< previous MAFD                           (scene detect only)

    AVFrame *srce[N_SRCE];              ///< buffered source frames
    int64_t srce_pts_dest[N_SRCE];      ///< pts for source frames scaled to output timebase
    double srce_score[N_SRCE];          ///< scene change score compared to the next srce frame
    int64_t pts;                        ///< pts of frame we are working on

    int max;
    int bitdepth;
    AVFrame *work;
} FrameRateContext;

#define OFFSET(x) offsetof(FrameRateContext, x)
#define V AV_OPT_FLAG_VIDEO_PARAM
#define F AV_OPT_FLAG_FILTERING_PARAM
#define FRAMERATE_FLAG_SCD 01

static const AVOption framerate_options[] = {
    {"fps",                 "required output frames per second rate", OFFSET(dest_frame_rate), AV_OPT_TYPE_VIDEO_RATE, {.str="50"},             0,       INT_MAX, V|F },

    {"interp_start",        "point to start linear interpolation",    OFFSET(interp_start),    AV_OPT_TYPE_INT,      {.i64=15},                 0,       255,     V|F },
    {"interp_end",          "point to end linear interpolation",      OFFSET(interp_end),      AV_OPT_TYPE_INT,      {.i64=240},                0,       255,     V|F },
    {"scene",               "scene change level",                     OFFSET(scene_score),     AV_OPT_TYPE_DOUBLE,   {.dbl=8.2},                0,       INT_MAX, V|F },

    {"flags",               "set flags",                              OFFSET(flags),           AV_OPT_TYPE_FLAGS,    {.i64=1},                  0,       INT_MAX, V|F, "flags" },
    {"scene_change_detect", "enable scene change detection",          0,                       AV_OPT_TYPE_CONST,    {.i64=FRAMERATE_FLAG_SCD}, INT_MIN, INT_MAX, V|F, "flags" },
    {"scd",                 "enable scene change detection",          0,                       AV_OPT_TYPE_CONST,    {.i64=FRAMERATE_FLAG_SCD}, INT_MIN, INT_MAX, V|F, "flags" },

    {NULL}
};

AVFILTER_DEFINE_CLASS(framerate);

static void next_source(AVFilterContext *ctx)
{
    FrameRateContext *s = ctx->priv;
    int i;

    ff_dlog(ctx,  "next_source()\n");

    if (s->srce[s->last] && s->srce[s->last] != s->srce[s->last-1]) {
        ff_dlog(ctx, "next_source() unlink %d\n", s->last);
        av_frame_free(&s->srce[s->last]);
    }
    for (i = s->last; i > s->frst; i--) {
        ff_dlog(ctx, "next_source() copy %d to %d\n", i - 1, i);
        s->srce[i] = s->srce[i - 1];
        s->srce_score[i] = s->srce_score[i - 1];
    }
    ff_dlog(ctx, "next_source() make %d null\n", s->frst);
    s->srce[s->frst] = NULL;
    s->srce_score[s->frst] = -1.0;
}

static av_always_inline int64_t sad_8x8_16(const uint16_t *src1, ptrdiff_t stride1,
                                           const uint16_t *src2, ptrdiff_t stride2)
{
    int sum = 0;
    int x, y;

    for (y = 0; y < 8; y++) {
        for (x = 0; x < 8; x++)
            sum += FFABS(src1[x] - src2[x]);
        src1 += stride1;
        src2 += stride2;
    }
    return sum;
}

static int64_t scene_sad16(FrameRateContext *s, const uint16_t *p1, int p1_linesize, const uint16_t* p2, int p2_linesize, const int width, const int height)
{
    int64_t sad;
    int x, y;
    for (sad = y = 0; y < height - 7; y += 8) {
        for (x = 0; x < width - 7; x += 8) {
            sad += sad_8x8_16(p1 + y * p1_linesize + x,
                              p1_linesize,
                              p2 + y * p2_linesize + x,
                              p2_linesize);
        }
    }
    return sad;
}

static int64_t scene_sad8(FrameRateContext *s, uint8_t *p1, int p1_linesize, uint8_t* p2, int p2_linesize, const int width, const int height)
{
    int64_t sad;
    int x, y;
    for (sad = y = 0; y < height - 7; y += 8) {
        for (x = 0; x < width - 7; x += 8) {
            sad += s->sad(p1 + y * p1_linesize + x,
                          p1_linesize,
                          p2 + y * p2_linesize + x,
                          p2_linesize);
        }
    }
    emms_c();
    return sad;
}

static double get_scene_score(AVFilterContext *ctx, AVFrame *crnt, AVFrame *next)
{
    FrameRateContext *s = ctx->priv;
    double ret = 0;

    ff_dlog(ctx, "get_scene_score()\n");

    if (crnt->height == next->height &&
        crnt->width  == next->width) {
        int64_t sad;
        double mafd, diff;

        ff_dlog(ctx, "get_scene_score() process\n");
        if (s->bitdepth == 8)
            sad = scene_sad8(s, crnt->data[0], crnt->linesize[0], next->data[0], next->linesize[0], crnt->width, crnt->height);
        else
            sad = scene_sad16(s, (const uint16_t*)crnt->data[0], crnt->linesize[0] / 2, (const uint16_t*)next->data[0], next->linesize[0] / 2, crnt->width, crnt->height);

        mafd = (double)sad * 100.0 / FFMAX(1, (crnt->height & ~7) * (crnt->width & ~7)) / (1 << s->bitdepth);
        diff = fabs(mafd - s->prev_mafd);
        ret  = av_clipf(FFMIN(mafd, diff), 0, 100.0);
        s->prev_mafd = mafd;
    }
    ff_dlog(ctx, "get_scene_score() result is:%f\n", ret);
    return ret;
}

typedef struct ThreadData {
    AVFrame *copy_src1, *copy_src2;
    uint16_t src1_factor, src2_factor;
} ThreadData;

static int filter_slice8(AVFilterContext *ctx, void *arg, int job, int nb_jobs)
{
    FrameRateContext *s = ctx->priv;
    ThreadData *td = arg;
    uint16_t src1_factor = td->src1_factor;
    uint16_t src2_factor = td->src2_factor;
    int plane, line, pixel;

    for (plane = 0; plane < 4 && td->copy_src1->data[plane] && td->copy_src2->data[plane]; plane++) {
        int cpy_line_width = s->line_size[plane];
        uint8_t *cpy_src1_data = td->copy_src1->data[plane];
        int cpy_src1_line_size = td->copy_src1->linesize[plane];
        uint8_t *cpy_src2_data = td->copy_src2->data[plane];
        int cpy_src2_line_size = td->copy_src2->linesize[plane];
        int cpy_src_h = (plane > 0 && plane < 3) ? (td->copy_src1->height >> s->vsub) : (td->copy_src1->height);
        uint8_t *cpy_dst_data = s->work->data[plane];
        int cpy_dst_line_size = s->work->linesize[plane];
        const int start = (cpy_src_h *  job   ) / nb_jobs;
        const int end   = (cpy_src_h * (job+1)) / nb_jobs;
        cpy_src1_data += start * cpy_src1_line_size;
        cpy_src2_data += start * cpy_src2_line_size;
        cpy_dst_data += start * cpy_dst_line_size;

        if (plane <1 || plane >2) {
            // luma or alpha
            for (line = start; line < end; line++) {
                for (pixel = 0; pixel < cpy_line_width; pixel++) {
                    // integer version of (src1 * src1_factor) + (src2 + src2_factor) + 0.5
                    // 0.5 is for rounding
                    // 128 is the integer representation of 0.5 << 8
                    cpy_dst_data[pixel] = ((cpy_src1_data[pixel] * src1_factor) + (cpy_src2_data[pixel] * src2_factor) + 128) >> 8;
                }
                cpy_src1_data += cpy_src1_line_size;
                cpy_src2_data += cpy_src2_line_size;
                cpy_dst_data += cpy_dst_line_size;
            }
        } else {
            // chroma
            for (line = start; line < end; line++) {
                for (pixel = 0; pixel < cpy_line_width; pixel++) {
                    // as above
                    // because U and V are based around 128 we have to subtract 128 from the components.
                    // 32896 is the integer representation of 128.5 << 8
                    cpy_dst_data[pixel] = (((cpy_src1_data[pixel] - 128) * src1_factor) + ((cpy_src2_data[pixel] - 128) * src2_factor) + 32896) >> 8;
                }
                cpy_src1_data += cpy_src1_line_size;
                cpy_src2_data += cpy_src2_line_size;
                cpy_dst_data += cpy_dst_line_size;
            }
        }
    }

    return 0;
}

static int filter_slice16(AVFilterContext *ctx, void *arg, int job, int nb_jobs)
{
    FrameRateContext *s = ctx->priv;
    ThreadData *td = arg;
    uint16_t src1_factor = td->src1_factor;
    uint16_t src2_factor = td->src2_factor;
    const int half = s->max / 2;
    const int uv = (s->max + 1) * half;
    const int shift = s->bitdepth;
    int plane, line, pixel;

    for (plane = 0; plane < 4 && td->copy_src1->data[plane] && td->copy_src2->data[plane]; plane++) {
        int cpy_line_width = s->line_size[plane];
        const uint16_t *cpy_src1_data = (const uint16_t *)td->copy_src1->data[plane];
        int cpy_src1_line_size = td->copy_src1->linesize[plane] / 2;
        const uint16_t *cpy_src2_data = (const uint16_t *)td->copy_src2->data[plane];
        int cpy_src2_line_size = td->copy_src2->linesize[plane] / 2;
        int cpy_src_h = (plane > 0 && plane < 3) ? (td->copy_src1->height >> s->vsub) : (td->copy_src1->height);
        uint16_t *cpy_dst_data = (uint16_t *)s->work->data[plane];
        int cpy_dst_line_size = s->work->linesize[plane] / 2;
        const int start = (cpy_src_h *  job   ) / nb_jobs;
        const int end   = (cpy_src_h * (job+1)) / nb_jobs;
        cpy_src1_data += start * cpy_src1_line_size;
        cpy_src2_data += start * cpy_src2_line_size;
        cpy_dst_data += start * cpy_dst_line_size;

        if (plane <1 || plane >2) {
            // luma or alpha
            for (line = start; line < end; line++) {
                for (pixel = 0; pixel < cpy_line_width; pixel++)
                    cpy_dst_data[pixel] = ((cpy_src1_data[pixel] * src1_factor) + (cpy_src2_data[pixel] * src2_factor) + half) >> shift;
                cpy_src1_data += cpy_src1_line_size;
                cpy_src2_data += cpy_src2_line_size;
                cpy_dst_data += cpy_dst_line_size;
            }
        } else {
            // chroma
            for (line = start; line < end; line++) {
                for (pixel = 0; pixel < cpy_line_width; pixel++) {
                    cpy_dst_data[pixel] = (((cpy_src1_data[pixel] - half) * src1_factor) + ((cpy_src2_data[pixel] - half) * src2_factor) + uv) >> shift;
                }
                cpy_src1_data += cpy_src1_line_size;
                cpy_src2_data += cpy_src2_line_size;
                cpy_dst_data += cpy_dst_line_size;
            }
        }
    }

    return 0;
}

static int blend_frames(AVFilterContext *ctx, float interpolate,
                        int src1, int src2)
{
    FrameRateContext *s = ctx->priv;
    AVFilterLink *outlink = ctx->outputs[0];
    double interpolate_scene_score = 0;

    if ((s->flags & FRAMERATE_FLAG_SCD) && s->srce[src1] && s->srce[src2]) {
        int i1 = src1 < src2 ? src1 : src2;
        int i2 = src1 < src2 ? src2 : src1;
        if (i2 == i1 + 1 && s->srce_score[i1] >= 0.0)
            interpolate_scene_score = s->srce_score[i1];
        else
            interpolate_scene_score = s->srce_score[i1] = get_scene_score(ctx, s->srce[i1], s->srce[i2]);
        ff_dlog(ctx, "blend_frames() interpolate scene score:%f\n", interpolate_scene_score);
    }
    // decide if the shot-change detection allows us to blend two frames
    if (interpolate_scene_score < s->scene_score && s->srce[src2]) {
        ThreadData td;
        td.copy_src1 = s->srce[src1];
        td.copy_src2 = s->srce[src2];
        td.src2_factor = fabsf(interpolate) * (1 << (s->bitdepth - 8));
        td.src1_factor = s->max - td.src2_factor;

        // get work-space for output frame
        s->work = ff_get_video_buffer(outlink, outlink->w, outlink->h);
        if (!s->work)
            return AVERROR(ENOMEM);

        av_frame_copy_props(s->work, s->srce[s->crnt]);

        ff_dlog(ctx, "blend_frames() INTERPOLATE to create work frame\n");
        ctx->internal->execute(ctx, s->bitdepth == 8 ? filter_slice8 : filter_slice16, &td, NULL, FFMIN(outlink->h, ff_filter_get_nb_threads(ctx)));
        return 1;
    }
    return 0;
}

static int process_work_frame(AVFilterContext *ctx, int stop)
{
    FrameRateContext *s = ctx->priv;
    int64_t work_next_pts;
    float interpolate;
    int src1, src2;

    ff_dlog(ctx, "process_work_frame()\n");

    ff_dlog(ctx, "process_work_frame() pending_input_frames %d\n", s->pending_srce_frames);

    if (s->srce[s->prev]) ff_dlog(ctx, "process_work_frame() srce prev pts:%"PRId64"\n", s->srce[s->prev]->pts);
    if (s->srce[s->crnt]) ff_dlog(ctx, "process_work_frame() srce crnt pts:%"PRId64"\n", s->srce[s->crnt]->pts);
    if (s->srce[s->next]) ff_dlog(ctx, "process_work_frame() srce next pts:%"PRId64"\n", s->srce[s->next]->pts);

    if (!s->srce[s->crnt]) {
        // the filter cannot do anything
        ff_dlog(ctx, "process_work_frame() no current frame cached: move on to next frame, do not output a frame\n");
        next_source(ctx);
        return 0;
    }

    work_next_pts = s->pts + s->average_dest_pts_delta;

    ff_dlog(ctx, "process_work_frame() work crnt pts:%"PRId64"\n", s->pts);
    ff_dlog(ctx, "process_work_frame() work next pts:%"PRId64"\n", work_next_pts);
    if (s->srce[s->prev])
        ff_dlog(ctx, "process_work_frame() srce prev pts:%"PRId64" at dest time base:%u/%u\n",
            s->srce_pts_dest[s->prev], s->dest_time_base.num, s->dest_time_base.den);
    if (s->srce[s->crnt])
        ff_dlog(ctx, "process_work_frame() srce crnt pts:%"PRId64" at dest time base:%u/%u\n",
            s->srce_pts_dest[s->crnt], s->dest_time_base.num, s->dest_time_base.den);
    if (s->srce[s->next])
        ff_dlog(ctx, "process_work_frame() srce next pts:%"PRId64" at dest time base:%u/%u\n",
            s->srce_pts_dest[s->next], s->dest_time_base.num, s->dest_time_base.den);

    av_assert0(s->srce[s->next]);

    // should filter be skipping input frame (output frame rate is lower than input frame rate)
    if (!s->flush && s->pts >= s->srce_pts_dest[s->next]) {
        ff_dlog(ctx, "process_work_frame() work crnt pts >= srce next pts: SKIP FRAME, move on to next frame, do not output a frame\n");
        next_source(ctx);
        s->pending_srce_frames--;
        return 0;
    }

    // calculate interpolation
    interpolate = ((s->pts - s->srce_pts_dest[s->crnt]) * 256.0 / s->average_srce_pts_dest_delta);
    ff_dlog(ctx, "process_work_frame() interpolate:%f/256\n", interpolate);
    src1 = s->crnt;
    if (interpolate > s->interp_end) {
        ff_dlog(ctx, "process_work_frame() source is:NEXT\n");
        src1 = s->next;
    }
    if (s->srce[s->prev] && interpolate < -s->interp_end) {
        ff_dlog(ctx, "process_work_frame() source is:PREV\n");
        src1 = s->prev;
    }

    // decide whether to blend two frames
    if ((interpolate >= s->interp_start && interpolate <= s->interp_end) || (interpolate <= -s->interp_start && interpolate >= -s->interp_end)) {
        if (interpolate > 0) {
            ff_dlog(ctx, "process_work_frame() interpolate source is:NEXT\n");
            src2 = s->next;
        } else {
            ff_dlog(ctx, "process_work_frame() interpolate source is:PREV\n");
            src2 = s->prev;
        }
        if (blend_frames(ctx, interpolate, src1, src2))
            goto copy_done;
        else
            ff_dlog(ctx, "process_work_frame() CUT - DON'T INTERPOLATE\n");
    }

    ff_dlog(ctx, "process_work_frame() COPY to the work frame\n");
    // copy the frame we decided is our base source
    s->work = av_frame_clone(s->srce[src1]);
    if (!s->work)
        return AVERROR(ENOMEM);

copy_done:
    s->work->pts = s->pts;

    // should filter be re-using input frame (output frame rate is higher than input frame rate)
    if (!s->flush && (work_next_pts + s->average_dest_pts_delta) < (s->srce_pts_dest[s->crnt] + s->average_srce_pts_dest_delta)) {
        ff_dlog(ctx, "process_work_frame() REPEAT FRAME\n");
    } else {
        ff_dlog(ctx, "process_work_frame() CONSUME FRAME, move to next frame\n");
        s->pending_srce_frames--;
        next_source(ctx);
    }
    ff_dlog(ctx, "process_work_frame() output a frame\n");
    s->dest_frame_num++;
    if (stop)
        s->pending_end_frame = 0;
    s->last_dest_frame_pts = s->work->pts;

    return 1;
}

static void set_srce_frame_dest_pts(AVFilterContext *ctx)
{
    FrameRateContext *s = ctx->priv;

    ff_dlog(ctx, "set_srce_frame_output_pts()\n");

    // scale the input pts from the timebase difference between input and output
    if (s->srce[s->prev])
        s->srce_pts_dest[s->prev] = av_rescale_q(s->srce[s->prev]->pts, s->srce_time_base, s->dest_time_base);
    if (s->srce[s->crnt])
        s->srce_pts_dest[s->crnt] = av_rescale_q(s->srce[s->crnt]->pts, s->srce_time_base, s->dest_time_base);
    if (s->srce[s->next])
        s->srce_pts_dest[s->next] = av_rescale_q(s->srce[s->next]->pts, s->srce_time_base, s->dest_time_base);
}

static void set_work_frame_pts(AVFilterContext *ctx)
{
    FrameRateContext *s = ctx->priv;
    int64_t pts, average_srce_pts_delta = 0;

    ff_dlog(ctx, "set_work_frame_pts()\n");

    av_assert0(s->srce[s->next]);
    av_assert0(s->srce[s->crnt]);

    ff_dlog(ctx, "set_work_frame_pts() srce crnt pts:%"PRId64"\n", s->srce[s->crnt]->pts);
    ff_dlog(ctx, "set_work_frame_pts() srce next pts:%"PRId64"\n", s->srce[s->next]->pts);
    if (s->srce[s->prev])
        ff_dlog(ctx, "set_work_frame_pts() srce prev pts:%"PRId64"\n", s->srce[s->prev]->pts);

    average_srce_pts_delta = s->average_srce_pts_dest_delta;
    ff_dlog(ctx, "set_work_frame_pts() initial average srce pts:%"PRId64"\n", average_srce_pts_delta);

    set_srce_frame_dest_pts(ctx);

    // calculate the PTS delta
    if ((pts = (s->srce_pts_dest[s->next] - s->srce_pts_dest[s->crnt]))) {
        average_srce_pts_delta = average_srce_pts_delta?((average_srce_pts_delta+pts)>>1):pts;
    } else if (s->srce[s->prev] && (pts = (s->srce_pts_dest[s->crnt] - s->srce_pts_dest[s->prev]))) {
        average_srce_pts_delta = average_srce_pts_delta?((average_srce_pts_delta+pts)>>1):pts;
    }

    s->average_srce_pts_dest_delta = average_srce_pts_delta;
    ff_dlog(ctx, "set_work_frame_pts() average srce pts:%"PRId64"\n", average_srce_pts_delta);
    ff_dlog(ctx, "set_work_frame_pts() average srce pts:%"PRId64" at dest time base:%u/%u\n",
            s->average_srce_pts_dest_delta, s->dest_time_base.num, s->dest_time_base.den);

    if (ctx->inputs[0] && !s->average_dest_pts_delta) {
        int64_t d = av_q2d(av_inv_q(av_mul_q(s->dest_time_base, s->dest_frame_rate)));
        s->average_dest_pts_delta = d;
        ff_dlog(ctx, "set_work_frame_pts() average dest pts delta:%"PRId64"\n", s->average_dest_pts_delta);
    }

    if (!s->dest_frame_num) {
        s->pts = s->last_dest_frame_pts = s->srce_pts_dest[s->crnt];
    } else {
        s->pts = s->last_dest_frame_pts + s->average_dest_pts_delta;
    }

    ff_dlog(ctx, "set_work_frame_pts() calculated pts:%"PRId64" at dest time base:%u/%u\n",
            s->pts, s->dest_time_base.num, s->dest_time_base.den);
}

static av_cold int init(AVFilterContext *ctx)
{
    FrameRateContext *s = ctx->priv;
    int i;

    s->dest_frame_num = 0;

    s->crnt = (N_SRCE)>>1;
    s->last = N_SRCE - 1;

    s->next = s->crnt - 1;
    s->prev = s->crnt + 1;

    for (i = 0; i < N_SRCE; i++)
        s->srce_score[i] = -1.0;

    return 0;
}

static av_cold void uninit(AVFilterContext *ctx)
{
    FrameRateContext *s = ctx->priv;
    int i;

    for (i = s->frst; i < s->last; i++) {
        if (s->srce[i] && (s->srce[i] != s->srce[i + 1]))
            av_frame_free(&s->srce[i]);
    }
    av_frame_free(&s->srce[s->last]);
}

static int query_formats(AVFilterContext *ctx)
{
    static const enum AVPixelFormat pix_fmts[] = {
        AV_PIX_FMT_YUV410P,
        AV_PIX_FMT_YUV411P, AV_PIX_FMT_YUVJ411P,
        AV_PIX_FMT_YUV420P, AV_PIX_FMT_YUVJ420P,
        AV_PIX_FMT_YUV422P, AV_PIX_FMT_YUVJ422P,
        AV_PIX_FMT_YUV440P, AV_PIX_FMT_YUVJ440P,
        AV_PIX_FMT_YUV444P, AV_PIX_FMT_YUVJ444P,
        AV_PIX_FMT_YUV420P9, AV_PIX_FMT_YUV420P10, AV_PIX_FMT_YUV420P12,
        AV_PIX_FMT_YUV422P9, AV_PIX_FMT_YUV422P10, AV_PIX_FMT_YUV422P12,
        AV_PIX_FMT_YUV444P9, AV_PIX_FMT_YUV444P10, AV_PIX_FMT_YUV444P12,
        AV_PIX_FMT_NONE
    };

    AVFilterFormats *fmts_list = ff_make_format_list(pix_fmts);
    if (!fmts_list)
        return AVERROR(ENOMEM);
    return ff_set_common_formats(ctx, fmts_list);
}

static int config_input(AVFilterLink *inlink)
{
    AVFilterContext *ctx = inlink->dst;
    FrameRateContext *s = ctx->priv;
    const AVPixFmtDescriptor *pix_desc = av_pix_fmt_desc_get(inlink->format);
    int plane;

    for (plane = 0; plane < 4; plane++) {
        s->line_size[plane] = av_image_get_linesize(inlink->format, inlink->w,
                                                    plane);
    }

    s->bitdepth = pix_desc->comp[0].depth;
    s->vsub = pix_desc->log2_chroma_h;

    s->sad = av_pixelutils_get_sad_fn(3, 3, 2, s); // 8x8 both sources aligned
    if (!s->sad)
        return AVERROR(EINVAL);

    s->srce_time_base = inlink->time_base;

    s->max = 1 << (s->bitdepth);

    return 0;
}

static int filter_frame(AVFilterLink *inlink, AVFrame *inpicref)
{
    int ret;
    AVFilterContext *ctx = inlink->dst;
    FrameRateContext *s = ctx->priv;

    // we have one new frame
    s->pending_srce_frames++;

    if (inpicref->interlaced_frame)
        av_log(ctx, AV_LOG_WARNING, "Interlaced frame found - the output will not be correct.\n");

    // store the pointer to the new frame
    av_frame_free(&s->srce[s->frst]);
    s->srce[s->frst] = inpicref;

    if (!s->pending_end_frame && s->srce[s->crnt]) {
        set_work_frame_pts(ctx);
        s->pending_end_frame = 1;
    } else {
        set_srce_frame_dest_pts(ctx);
    }

    ret = process_work_frame(ctx, 1);
    if (ret < 0)
        return ret;
    return ret ? ff_filter_frame(ctx->outputs[0], s->work) : 0;
}

static int config_output(AVFilterLink *outlink)
{
    AVFilterContext *ctx = outlink->src;
    FrameRateContext *s = ctx->priv;
    int exact;

    ff_dlog(ctx, "config_output()\n");

    ff_dlog(ctx,
           "config_output() input time base:%u/%u (%f)\n",
           ctx->inputs[0]->time_base.num,ctx->inputs[0]->time_base.den,
           av_q2d(ctx->inputs[0]->time_base));

    // make sure timebase is small enough to hold the framerate

    exact = av_reduce(&s->dest_time_base.num, &s->dest_time_base.den,
                      av_gcd((int64_t)s->srce_time_base.num * s->dest_frame_rate.num,
                             (int64_t)s->srce_time_base.den * s->dest_frame_rate.den ),
                      (int64_t)s->srce_time_base.den * s->dest_frame_rate.num, INT_MAX);

    av_log(ctx, AV_LOG_INFO,
           "time base:%u/%u -> %u/%u exact:%d\n",
           s->srce_time_base.num, s->srce_time_base.den,
           s->dest_time_base.num, s->dest_time_base.den, exact);
    if (!exact) {
        av_log(ctx, AV_LOG_WARNING, "Timebase conversion is not exact\n");
    }

    outlink->frame_rate = s->dest_frame_rate;
    outlink->time_base = s->dest_time_base;

    ff_dlog(ctx,
           "config_output() output time base:%u/%u (%f) w:%d h:%d\n",
           outlink->time_base.num, outlink->time_base.den,
           av_q2d(outlink->time_base),
           outlink->w, outlink->h);


    av_log(ctx, AV_LOG_INFO, "fps -> fps:%u/%u scene score:%f interpolate start:%d end:%d\n",
            s->dest_frame_rate.num, s->dest_frame_rate.den,
            s->scene_score, s->interp_start, s->interp_end);

    return 0;
}

static int request_frame(AVFilterLink *outlink)
{
    AVFilterContext *ctx = outlink->src;
    FrameRateContext *s = ctx->priv;
    int ret, i;

    ff_dlog(ctx, "request_frame()\n");

    // if there is no "next" frame AND we are not in flush then get one from our input filter
    if (!s->srce[s->frst] && !s->flush)
        goto request;

    ff_dlog(ctx, "request_frame() REPEAT or FLUSH\n");

    if (s->pending_srce_frames <= 0) {
        ff_dlog(ctx, "request_frame() nothing else to do, return:EOF\n");
        return AVERROR_EOF;
    }

    // otherwise, make brand-new frame and pass to our output filter
    ff_dlog(ctx, "request_frame() FLUSH\n");

    // back fill at end of file when source has no more frames
    for (i = s->last; i > s->frst; i--) {
        if (!s->srce[i - 1] && s->srce[i]) {
            ff_dlog(ctx, "request_frame() copy:%d to:%d\n", i, i - 1);
            s->srce[i - 1] = s->srce[i];
        }
    }

    set_work_frame_pts(ctx);
    ret = process_work_frame(ctx, 0);
    if (ret < 0)
        return ret;
    if (ret)
        return ff_filter_frame(ctx->outputs[0], s->work);

request:
    ff_dlog(ctx, "request_frame() call source's request_frame()\n");
    ret = ff_request_frame(ctx->inputs[0]);
    if (ret < 0 && (ret != AVERROR_EOF)) {
        ff_dlog(ctx, "request_frame() source's request_frame() returned error:%d\n", ret);
        return ret;
    } else if (ret == AVERROR_EOF) {
        s->flush = 1;
    }
    ff_dlog(ctx, "request_frame() source's request_frame() returned:%d\n", ret);
    return 0;
}

static const AVFilterPad framerate_inputs[] = {
    {
        .name         = "default",
        .type         = AVMEDIA_TYPE_VIDEO,
        .config_props = config_input,
        .filter_frame = filter_frame,
    },
    { NULL }
};

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

AVFilter ff_vf_framerate = {
    .name          = "framerate",
    .description   = NULL_IF_CONFIG_SMALL("Upsamples or downsamples progressive source between specified frame rates."),
    .priv_size     = sizeof(FrameRateContext),
    .priv_class    = &framerate_class,
    .init          = init,
    .uninit        = uninit,
    .query_formats = query_formats,
    .inputs        = framerate_inputs,
    .outputs       = framerate_outputs,
    .flags         = AVFILTER_FLAG_SLICE_THREADS,
};