From patchwork Mon Nov 19 13:46:08 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Paul B Mahol X-Patchwork-Id: 11074 Return-Path: X-Original-To: patchwork@ffaux-bg.ffmpeg.org Delivered-To: patchwork@ffaux-bg.ffmpeg.org Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org [79.124.17.100]) by ffaux.localdomain (Postfix) with ESMTP id 2397644D2A4 for ; Mon, 19 Nov 2018 15:52:53 +0200 (EET) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 8F524689DE9; Mon, 19 Nov 2018 15:52:53 +0200 (EET) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-wm1-f66.google.com (mail-wm1-f66.google.com [209.85.128.66]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 52CE9689CF2 for ; Mon, 19 Nov 2018 15:52:47 +0200 (EET) Received: by mail-wm1-f66.google.com with SMTP id w7-v6so5405676wmc.1 for ; Mon, 19 Nov 2018 05:52:49 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:subject:date:message-id; bh=oJ44fQURe7+VYxuQb435QEoWP4h4BkIZ72r22GNTci0=; b=P7xhZ62c1YigyHP7yKRlUBK8ZYbZvS0UGPvdREnQ/c5dREhJ7vyrKtjQ5hAUAKPCwZ cgA5HLGooGw7hG6u0kv6sFTLEf1pk3MRllRqMob2CQQiGtfJ8AOzV5L/WUYCFd0V4NuJ VI9S6fQ6vJo5xUF1xGVAmKMMYYKFIW4Bkw7BaHJjnD/SMo5v0BrmKGAHIfvtzQ+HjBEO pkSKf8dc6dCcmPzLbDuNuVkiQfsBg/i1By9BAsWhNLOw41PD29BmDfJ5phXw8r4RPomk Ts3Fbt+JkvwmmmMta0mw/DRosO3DxgnUwLjU77NJHo7tDeHDrjRlcHheGm9ueIH+y05c 9Cyg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:subject:date:message-id; bh=oJ44fQURe7+VYxuQb435QEoWP4h4BkIZ72r22GNTci0=; b=qxdMSyigT2/hrICQxMZgOK8MTATB3Mc0PBPg3p+DKhrNuEVx//U89c92yNPtHz3VFm jY4yl8MZ3+I8Xa/otLe1a243uQ91CX2UIvvVeJlqY5ZazBW850ui7zC1O0zr9uVHl9xX jM2f8g6KhAexDkt3zE7agAsqmNdxd4hRxoVaG49P2mRglZL7Y7wpfQES0cz8V1eQxIkj ovb0SBnNaq/yrJbkrJwjlzpa65mWh2hdvZNir341ezQdWABJSU52IEleHT/OodB8Eb41 3WfgXyY5dPPSqcsTvoJe+XGADMSQsRsqFhMt4DHb8eAkfCEG1uKPBQ/O1VE9horDC+JO VmFQ== X-Gm-Message-State: AGRZ1gKF3wjdsmFsnBklR1JzCoq9FnTw2aNy5zTm5JjqTImZ7sM496+B ijJ7KigfdcARnRifiDC01OSoPunk X-Google-Smtp-Source: AFSGD/XJhZa466kSWnuw9iPoL1h0MVjD1hD8b3S8NeAh6Tg3C3S5zhwszA30aCzlFjYDm72gTO7gdw== X-Received: by 2002:a1c:28c6:: with SMTP id o189-v6mr7002065wmo.89.1542635178383; Mon, 19 Nov 2018 05:46:18 -0800 (PST) Received: from localhost.localdomain ([94.250.174.60]) by smtp.gmail.com with ESMTPSA id m15sm10023419wrr.95.2018.11.19.05.46.16 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 19 Nov 2018 05:46:17 -0800 (PST) From: Paul B Mahol To: ffmpeg-devel@ffmpeg.org Date: Mon, 19 Nov 2018 14:46:08 +0100 Message-Id: <20181119134608.8797-1-onemda@gmail.com> X-Mailer: git-send-email 2.17.1 Subject: [FFmpeg-devel] [PATCH] avfilter: add maskfun filter X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: FFmpeg development discussions and patches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Reply-To: FFmpeg development discussions and patches MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" Signed-off-by: Paul B Mahol --- doc/filters.texi | 27 ++++ libavfilter/Makefile | 1 + libavfilter/allfilters.c | 1 + libavfilter/vf_maskfun.c | 279 +++++++++++++++++++++++++++++++++++++++ 4 files changed, 308 insertions(+) create mode 100644 libavfilter/vf_maskfun.c diff --git a/doc/filters.texi b/doc/filters.texi index a697d3db14..0cbd2ac69c 100644 --- a/doc/filters.texi +++ b/doc/filters.texi @@ -11808,6 +11808,33 @@ copied from first stream. By default value 0xf, all planes will be processed. @end table +@section maskfun +Create mask from input video. + +For example it is useful to create motion masks after @code{tblend} filter. + +This filter accepts the following options: + +@table @option +@item low +Set low threshold. Any pixel component lower or exact than this value will be set to 0. + +@item high +Set high threshold. Any pixel component higher than this value will be set to max value +allowed for current pixel format. + +@item planes +Set planes to filter, by default all available planes are filtered. + +@item fill +Fill all frame pixels with this value. + +@item sum +Set max average pixel value for frame. If sum of all pixel components is higher that this +average, output frame will be completely filled with value set by @var{fill} option. +Typically useful for scene changes when used in combination with @code{tblend} filter. +@end table + @section mcdeint Apply motion-compensation deinterlacing. diff --git a/libavfilter/Makefile b/libavfilter/Makefile index 6058fb97bc..fe2f42e360 100644 --- a/libavfilter/Makefile +++ b/libavfilter/Makefile @@ -277,6 +277,7 @@ OBJS-$(CONFIG_LUTRGB_FILTER) += vf_lut.o OBJS-$(CONFIG_LUTYUV_FILTER) += vf_lut.o OBJS-$(CONFIG_MASKEDCLAMP_FILTER) += vf_maskedclamp.o framesync.o OBJS-$(CONFIG_MASKEDMERGE_FILTER) += vf_maskedmerge.o framesync.o +OBJS-$(CONFIG_MASKFUN_FILTER) += vf_maskfun.o OBJS-$(CONFIG_MCDEINT_FILTER) += vf_mcdeint.o OBJS-$(CONFIG_MERGEPLANES_FILTER) += vf_mergeplanes.o framesync.o OBJS-$(CONFIG_MESTIMATE_FILTER) += vf_mestimate.o motion_estimation.o diff --git a/libavfilter/allfilters.c b/libavfilter/allfilters.c index 8dfa572fde..7cd2f187fe 100644 --- a/libavfilter/allfilters.c +++ b/libavfilter/allfilters.c @@ -263,6 +263,7 @@ extern AVFilter ff_vf_lutrgb; extern AVFilter ff_vf_lutyuv; extern AVFilter ff_vf_maskedclamp; extern AVFilter ff_vf_maskedmerge; +extern AVFilter ff_vf_maskfun; extern AVFilter ff_vf_mcdeint; extern AVFilter ff_vf_mergeplanes; extern AVFilter ff_vf_mestimate; diff --git a/libavfilter/vf_maskfun.c b/libavfilter/vf_maskfun.c new file mode 100644 index 0000000000..a8c6466d22 --- /dev/null +++ b/libavfilter/vf_maskfun.c @@ -0,0 +1,279 @@ +/* + * Copyright (c) 2018 Paul B Mahol + * + * 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 + */ + +#include "libavutil/imgutils.h" +#include "libavutil/pixdesc.h" +#include "libavutil/opt.h" +#include "avfilter.h" +#include "formats.h" +#include "internal.h" +#include "video.h" + +typedef struct MaskFunContext { + const AVClass *class; + + int low, high; + int planes; + int fill; + int sum; + + int linesize[4]; + int width[4], height[4]; + int nb_planes; + int depth; + int max; + uint64_t max_sum; + + AVFrame *empty; + int (*getsum)(AVFilterContext *ctx, AVFrame *out); + int (*maskfun)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs); +} MaskFunContext; + +#define OFFSET(x) offsetof(MaskFunContext, x) +#define VF AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM + +static const AVOption maskfun_options[] = { + { "low", "set low threshold", OFFSET(low), AV_OPT_TYPE_INT, {.i64=10}, 0, UINT16_MAX, VF }, + { "high", "set high threshold", OFFSET(high), AV_OPT_TYPE_INT, {.i64=10}, 0, UINT16_MAX, VF }, + { "planes", "set planes", OFFSET(planes), AV_OPT_TYPE_INT, {.i64=0xF}, 0, 0xF, VF }, + { "fill", "set fill value", OFFSET(fill), AV_OPT_TYPE_INT, {.i64=0}, 0, UINT16_MAX, VF }, + { "sum", "set sum value", OFFSET(sum), AV_OPT_TYPE_INT, {.i64=10}, 0, UINT16_MAX, VF }, + { NULL } +}; + +AVFILTER_DEFINE_CLASS(maskfun); + +static int query_formats(AVFilterContext *ctx) +{ + static const enum AVPixelFormat pix_fmts[] = { + AV_PIX_FMT_YUVA444P, AV_PIX_FMT_YUV444P, AV_PIX_FMT_YUV440P, + AV_PIX_FMT_YUVJ444P, AV_PIX_FMT_YUVJ440P, + AV_PIX_FMT_YUVA422P, AV_PIX_FMT_YUV422P, AV_PIX_FMT_YUVA420P, AV_PIX_FMT_YUV420P, + AV_PIX_FMT_YUVJ422P, AV_PIX_FMT_YUVJ420P, + AV_PIX_FMT_YUVJ411P, AV_PIX_FMT_YUV411P, AV_PIX_FMT_YUV410P, + AV_PIX_FMT_YUV420P9, AV_PIX_FMT_YUV422P9, AV_PIX_FMT_YUV444P9, + AV_PIX_FMT_YUV420P10, AV_PIX_FMT_YUV422P10, AV_PIX_FMT_YUV444P10, + AV_PIX_FMT_YUV420P12, AV_PIX_FMT_YUV422P12, AV_PIX_FMT_YUV444P12, AV_PIX_FMT_YUV440P12, + AV_PIX_FMT_YUV420P14, AV_PIX_FMT_YUV422P14, AV_PIX_FMT_YUV444P14, + AV_PIX_FMT_YUV420P16, AV_PIX_FMT_YUV422P16, AV_PIX_FMT_YUV444P16, + AV_PIX_FMT_YUVA420P9, AV_PIX_FMT_YUVA422P9, AV_PIX_FMT_YUVA444P9, + AV_PIX_FMT_YUVA420P10, AV_PIX_FMT_YUVA422P10, AV_PIX_FMT_YUVA444P10, + AV_PIX_FMT_YUVA420P16, AV_PIX_FMT_YUVA422P16, AV_PIX_FMT_YUVA444P16, + AV_PIX_FMT_GBRP, AV_PIX_FMT_GBRP9, AV_PIX_FMT_GBRP10, + AV_PIX_FMT_GBRP12, AV_PIX_FMT_GBRP14, AV_PIX_FMT_GBRP16, + AV_PIX_FMT_GBRAP, AV_PIX_FMT_GBRAP10, AV_PIX_FMT_GBRAP12, AV_PIX_FMT_GBRAP16, + AV_PIX_FMT_GRAY8, AV_PIX_FMT_GRAY9, AV_PIX_FMT_GRAY10, AV_PIX_FMT_GRAY12, AV_PIX_FMT_GRAY14, AV_PIX_FMT_GRAY16, + AV_PIX_FMT_NONE + }; + + return ff_set_common_formats(ctx, ff_make_format_list(pix_fmts)); +} + +static int filter_frame(AVFilterLink *inlink, AVFrame *frame) +{ + AVFilterContext *ctx = inlink->dst; + MaskFunContext *s = ctx->priv; + AVFilterLink *outlink = ctx->outputs[0]; + + if (s->getsum(ctx, frame)) { + AVFrame *out = av_frame_clone(s->empty); + + if (!out) { + av_frame_free(&frame); + return AVERROR(ENOMEM); + } + out->pts = frame->pts; + av_frame_free(&frame); + + return ff_filter_frame(outlink, out); + } + + ctx->internal->execute(ctx, s->maskfun, frame, NULL, + FFMIN(s->height[1], ff_filter_get_nb_threads(ctx))); + + return ff_filter_frame(outlink, frame); +} + +#define GETSUM(name, type, div) \ +static int getsum##name(AVFilterContext *ctx, AVFrame *out) \ +{ \ + MaskFunContext *s = ctx->priv; \ + uint64_t sum = 0; \ + int p; \ + \ + for (p = 0; p < s->nb_planes; p++) { \ + const int linesize = out->linesize[p] / div; \ + const int w = s->width[p]; \ + const int h = s->height[p]; \ + type *dst = (type *)out->data[p]; \ + \ + if (!((1 << p) & s->planes)) \ + continue; \ + \ + for (int y = 0; y < h; y++) { \ + for (int x = 0; x < w; x++) \ + sum += dst[x]; \ + if (sum >= s->max_sum) \ + return 1; \ + dst += linesize; \ + } \ + } \ + \ + return 0; \ +} + +GETSUM(8, uint8_t, 1) +GETSUM(16, uint16_t, 2) + +#define MASKFUN(name, type, div) \ +static int maskfun##name(AVFilterContext *ctx, void *arg, \ + int jobnr, int nb_jobs) \ +{ \ + MaskFunContext *s = ctx->priv; \ + AVFrame *out = arg; \ + const int low = s->low; \ + const int high = s->high; \ + const int max = s->max; \ + int p; \ + \ + for (p = 0; p < s->nb_planes; p++) { \ + const int linesize = out->linesize[p] / div; \ + const int w = s->width[p]; \ + const int h = s->height[p]; \ + const int slice_start = (h * jobnr) / nb_jobs; \ + const int slice_end = (h * (jobnr+1)) / nb_jobs; \ + type *dst = (type *)out->data[p] + slice_start * linesize; \ + \ + if (!((1 << p) & s->planes)) \ + continue; \ + \ + for (int y = slice_start; y < slice_end; y++) { \ + for (int x = 0; x < w; x++) { \ + if (dst[x] <= low) \ + dst[x] = 0; \ + else if (dst[x] > high) \ + dst[x] = max; \ + } \ + \ + dst += linesize; \ + } \ + } \ + \ + return 0; \ +} + +MASKFUN(8, uint8_t, 1) +MASKFUN(16, uint16_t, 2) + +static int config_input(AVFilterLink *inlink) +{ + AVFilterContext *ctx = inlink->dst; + MaskFunContext *s = ctx->priv; + const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(inlink->format); + int vsub, hsub, ret; + + s->nb_planes = av_pix_fmt_count_planes(inlink->format); + + if ((ret = av_image_fill_linesizes(s->linesize, inlink->format, inlink->w)) < 0) + return ret; + + hsub = desc->log2_chroma_w; + vsub = desc->log2_chroma_h; + s->height[1] = s->height[2] = AV_CEIL_RSHIFT(inlink->h, vsub); + s->height[0] = s->height[3] = inlink->h; + s->width[1] = s->width[2] = AV_CEIL_RSHIFT(inlink->w, hsub); + s->width[0] = s->width[3] = inlink->w; + + s->depth = desc->comp[0].depth; + s->max = (1 << s->depth) - 1; + s->fill = FFMIN(s->fill, s->max); + + if (s->depth == 8) { + s->maskfun = maskfun8; + s->getsum = getsum8; + } else { + s->maskfun = maskfun16; + s->getsum = getsum16; + } + + s->empty = ff_get_video_buffer(inlink, inlink->w, inlink->h); + if (!s->empty) + return AVERROR(ENOMEM); + + if (s->depth == 8) { + for (int p = 0; p < s->nb_planes; p++) { + uint8_t *dst = s->empty->data[p]; + + for (int y = 0; y < s->height[p]; y++) { + memset(dst, s->fill, s->width[p]); + dst += s->empty->linesize[p]; + } + } + } else { + for (int p = 0; p < s->nb_planes; p++) { + uint16_t *dst = (uint16_t *)s->empty->data[p]; + + for (int y = 0; y < s->height[p]; y++) { + for (int x = 0; x < s->width[p]; x++) + dst[x] = s->fill; + dst += s->empty->linesize[p] / 2; + } + } + } + + s->max_sum = 0; + for (int p = 0; p < s->nb_planes; p++) { + if (!((1 << p) & s->planes)) + continue; + s->max_sum += (uint64_t)s->sum * s->width[p] * s->height[p]; + } + + return 0; +} + +static const AVFilterPad maskfun_inputs[] = { + { + .name = "default", + .type = AVMEDIA_TYPE_VIDEO, + .filter_frame = filter_frame, + .config_props = config_input, + .needs_writable = 1, + }, + { NULL } +}; + +static const AVFilterPad maskfun_outputs[] = { + { + .name = "default", + .type = AVMEDIA_TYPE_VIDEO, + }, + { NULL } +}; + +AVFilter ff_vf_maskfun = { + .name = "maskfun", + .description = NULL_IF_CONFIG_SMALL("Create Mask."), + .priv_size = sizeof(MaskFunContext), + .query_formats = query_formats, + .inputs = maskfun_inputs, + .outputs = maskfun_outputs, + .priv_class = &maskfun_class, + .flags = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC | AVFILTER_FLAG_SLICE_THREADS, +};