October 2021 - ffmpeg-cvslog - lists.ffmpeg.org

avfilter/vf_dblur: add clipping
by Paul B Mahol 24 Oct '21

24 Oct '21

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Sat Oct 23 23:51:42 2021 +0200| [97cff84f106ff348a48915866ba7f487a77f2c31] | committer: Paul B Mahol avfilter/vf_dblur: add clipping > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=97cff84f106ff348a… --- libavfilter/vf_dblur.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavfilter/vf_dblur.c b/libavfilter/vf_dblur.c index c1e20fef6c..e821029906 100644 --- a/libavfilter/vf_dblur.c +++ b/libavfilter/vf_dblur.c @@ -241,7 +241,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in) if (s->depth == 8) { for (y = 0; y < height; y++) { for (x = 0; x < width; x++) { - dst[x] = lrintf(bptr[x]); + dst[x] = av_clip_uint8(lrintf(bptr[x])); } bptr += width; dst += out->linesize[plane]; @@ -249,7 +249,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in) } else { for (y = 0; y < height; y++) { for (x = 0; x < width; x++) { - dst16[x] = lrintf(bptr[x]); + dst16[x] = av_clip_uintp2_c(lrintf(bptr[x]), s->depth); } bptr += width; dst16 += out->linesize[plane] / 2;

1 0

avfilter/vf_dblur: use lrintf()
by Paul B Mahol 24 Oct '21

24 Oct '21

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Sat Oct 23 20:31:24 2021 +0200| [2148934ce3145940419ccc5b2d7ae93d11a043aa] | committer: Paul B Mahol avfilter/vf_dblur: use lrintf() > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=2148934ce31459404… --- libavfilter/vf_dblur.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavfilter/vf_dblur.c b/libavfilter/vf_dblur.c index 48c21c81b6..c1e20fef6c 100644 --- a/libavfilter/vf_dblur.c +++ b/libavfilter/vf_dblur.c @@ -241,7 +241,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in) if (s->depth == 8) { for (y = 0; y < height; y++) { for (x = 0; x < width; x++) { - dst[x] = bptr[x]; + dst[x] = lrintf(bptr[x]); } bptr += width; dst += out->linesize[plane]; @@ -249,7 +249,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in) } else { for (y = 0; y < height; y++) { for (x = 0; x < width; x++) { - dst16[x] = bptr[x]; + dst16[x] = lrintf(bptr[x]); } bptr += width; dst16 += out->linesize[plane] / 2;

1 0

avfilter/vf_dblur: fix artifacts
by Paul B Mahol 24 Oct '21

24 Oct '21

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Sat Oct 23 16:59:38 2021 +0200| [a586ce0a21a29503e318479aa965fdb3eb5d98f7] | committer: Paul B Mahol avfilter/vf_dblur: fix artifacts > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=a586ce0a21a29503e… --- libavfilter/vf_dblur.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavfilter/vf_dblur.c b/libavfilter/vf_dblur.c index 8ea1229629..48c21c81b6 100644 --- a/libavfilter/vf_dblur.c +++ b/libavfilter/vf_dblur.c @@ -68,7 +68,7 @@ static int filter_horizontally(AVFilterContext *ctx, int width, int height) if (s->R3 > 0) { for (int y = 1; y < height - 1; y++) { - g = q * f(0, 0) + c * f(0, 0); + g = q * f(y, 0) + c * f(y, 0); for (int x = 0; x < width; x++) { f(y, x) = b0 * f(y, x) + b1 * f(y - 1, x) + g; g = q * f(y, x) + c * f(y - 1, x); @@ -84,7 +84,7 @@ static int filter_horizontally(AVFilterContext *ctx, int width, int height) } } else { for (int y = 1; y < height - 1; y++) { - g = q * f(0, width - 1) + c * f(0, width - 1); + g = q * f(y, width - 1) + c * f(y, width - 1); for (int x = width - 1; x >= 0; x--) { f(y, x) = b0 * f(y, x) + b1 * f(y - 1, x) + g; g = q * f(y, x) + c * f(y - 1, x);

1 0

avfilter/vf_convolution: improve runtime support for convolution filter
by Paul B Mahol 23 Oct '21

23 Oct '21

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Sat Oct 23 14:47:11 2021 +0200| [6c45d34e505f7ef93e07db4a616e7b98d1490102] | committer: Paul B Mahol avfilter/vf_convolution: improve runtime support for convolution filter > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=6c45d34e505f7ef93… --- libavfilter/vf_convolution.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/libavfilter/vf_convolution.c b/libavfilter/vf_convolution.c index 47b59de421..6e6fe0d887 100644 --- a/libavfilter/vf_convolution.c +++ b/libavfilter/vf_convolution.c @@ -709,12 +709,14 @@ static int param_init(AVFilterContext *ctx) if (!strcmp(ctx->filter->name, "convolution")) { for (i = 0; i < 4; i++) { int *matrix = (int *)s->matrix[i]; - char *p, *arg, *saveptr = NULL; - float sum = 0; + char *orig, *p, *arg, *saveptr = NULL; + float sum = 1.f; - p = s->matrix_str[i]; + p = orig = av_strdup(s->matrix_str[i]); if (p) { s->matrix_length[i] = 0; + s->rdiv[i] = 0.f; + sum = 0.f; while (s->matrix_length[i] < 49) { if (!(arg = av_strtok(p, " |", &saveptr))) @@ -726,6 +728,7 @@ static int param_init(AVFilterContext *ctx) s->matrix_length[i]++; } + av_freep(&orig); if (!(s->matrix_length[i] & 1)) { av_log(ctx, AV_LOG_ERROR, "number of matrix elements must be odd\n"); return AVERROR(EINVAL);

1 0

avfilter/vf_convolution: unbreak runtime planes option functionality
by Paul B Mahol 23 Oct '21

23 Oct '21

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Sat Oct 23 14:15:24 2021 +0200| [9e8bb9efd9f06d6bb879f0c07dd48ea5c33ac376] | committer: Paul B Mahol avfilter/vf_convolution: unbreak runtime planes option functionality For non-convolution filters like sobel. > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=9e8bb9efd9f06d6bb… --- libavfilter/vf_convolution.c | 210 ++++++++++++++++++++----------------------- 1 file changed, 96 insertions(+), 114 deletions(-) diff --git a/libavfilter/vf_convolution.c b/libavfilter/vf_convolution.c index cc9ba2399e..47b59de421 100644 --- a/libavfilter/vf_convolution.c +++ b/libavfilter/vf_convolution.c @@ -699,96 +699,12 @@ static int filter_slice(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs) return 0; } -static int config_input(AVFilterLink *inlink) +static int param_init(AVFilterContext *ctx) { - AVFilterContext *ctx = inlink->dst; ConvolutionContext *s = ctx->priv; + AVFilterLink *inlink = ctx->inputs[0]; const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(inlink->format); - int p; - - s->depth = desc->comp[0].depth; - s->max = (1 << s->depth) - 1; - - s->planewidth[1] = s->planewidth[2] = AV_CEIL_RSHIFT(inlink->w, desc->log2_chroma_w); - s->planewidth[0] = s->planewidth[3] = inlink->w; - s->planeheight[1] = s->planeheight[2] = AV_CEIL_RSHIFT(inlink->h, desc->log2_chroma_h); - s->planeheight[0] = s->planeheight[3] = inlink->h; - - s->nb_planes = av_pix_fmt_count_planes(inlink->format); - s->nb_threads = ff_filter_get_nb_threads(ctx); - s->bpc = (s->depth + 7) / 8; - - if (!strcmp(ctx->filter->name, "convolution")) { - if (s->depth > 8) { - for (p = 0; p < s->nb_planes; p++) { - if (s->mode[p] == MATRIX_ROW) - s->filter[p] = filter16_row; - else if (s->mode[p] == MATRIX_COLUMN) - s->filter[p] = filter16_column; - else if (s->size[p] == 3) - s->filter[p] = filter16_3x3; - else if (s->size[p] == 5) - s->filter[p] = filter16_5x5; - else if (s->size[p] == 7) - s->filter[p] = filter16_7x7; - } - } -#if CONFIG_CONVOLUTION_FILTER && ARCH_X86_64 - ff_convolution_init_x86(s); -#endif - } else if (!strcmp(ctx->filter->name, "prewitt")) { - if (s->depth > 8) - for (p = 0; p < s->nb_planes; p++) - s->filter[p] = filter16_prewitt; - } else if (!strcmp(ctx->filter->name, "roberts")) { - if (s->depth > 8) - for (p = 0; p < s->nb_planes; p++) - s->filter[p] = filter16_roberts; - } else if (!strcmp(ctx->filter->name, "sobel")) { - if (s->depth > 8) - for (p = 0; p < s->nb_planes; p++) - s->filter[p] = filter16_sobel; - } else if (!strcmp(ctx->filter->name, "kirsch")) { - if (s->depth > 8) - for (p = 0; p < s->nb_planes; p++) - s->filter[p] = filter16_kirsch; - } else if (!strcmp(ctx->filter->name, "scharr")) { - if (s->depth > 8) - for (p = 0; p < s->nb_planes; p++) - s->filter[p] = filter16_scharr; - } - - return 0; -} - -static int filter_frame(AVFilterLink *inlink, AVFrame *in) -{ - AVFilterContext *ctx = inlink->dst; - ConvolutionContext *s = ctx->priv; - AVFilterLink *outlink = ctx->outputs[0]; - AVFrame *out; - ThreadData td; - - out = ff_get_video_buffer(outlink, outlink->w, outlink->h); - if (!out) { - av_frame_free(&in); - return AVERROR(ENOMEM); - } - av_frame_copy_props(out, in); - - td.in = in; - td.out = out; - ff_filter_execute(ctx, filter_slice, &td, NULL, - FFMIN3(s->planeheight[1], s->planewidth[1], s->nb_threads)); - - av_frame_free(&in); - return ff_filter_frame(outlink, out); -} - -static av_cold int init(AVFilterContext *ctx) -{ - ConvolutionContext *s = ctx->priv; - int i; + int p, i; if (!strcmp(ctx->filter->name, "convolution")) { for (i = 0; i < 4; i++) { @@ -866,10 +782,8 @@ static av_cold int init(AVFilterContext *ctx) } } else if (!strcmp(ctx->filter->name, "prewitt")) { for (i = 0; i < 4; i++) { - if ((1 << i) & s->planes) - s->filter[i] = filter_prewitt; - else - s->copy[i] = 1; + s->filter[i] = filter_prewitt; + s->copy[i] = !((1 << i) & s->planes); s->size[i] = 3; s->setup[i] = setup_3x3; s->rdiv[i] = s->scale; @@ -877,10 +791,8 @@ static av_cold int init(AVFilterContext *ctx) } } else if (!strcmp(ctx->filter->name, "roberts")) { for (i = 0; i < 4; i++) { - if ((1 << i) & s->planes) - s->filter[i] = filter_roberts; - else - s->copy[i] = 1; + s->filter[i] = filter_roberts; + s->copy[i] = !((1 << i) & s->planes); s->size[i] = 3; s->setup[i] = setup_3x3; s->rdiv[i] = s->scale; @@ -888,10 +800,8 @@ static av_cold int init(AVFilterContext *ctx) } } else if (!strcmp(ctx->filter->name, "sobel")) { for (i = 0; i < 4; i++) { - if ((1 << i) & s->planes) - s->filter[i] = filter_sobel; - else - s->copy[i] = 1; + s->filter[i] = filter_sobel; + s->copy[i] = !((1 << i) & s->planes); s->size[i] = 3; s->setup[i] = setup_3x3; s->rdiv[i] = s->scale; @@ -899,10 +809,8 @@ static av_cold int init(AVFilterContext *ctx) } } else if (!strcmp(ctx->filter->name, "kirsch")) { for (i = 0; i < 4; i++) { - if ((1 << i) & s->planes) - s->filter[i] = filter_kirsch; - else - s->copy[i] = 1; + s->filter[i] = filter_kirsch; + s->copy[i] = !((1 << i) & s->planes); s->size[i] = 3; s->setup[i] = setup_3x3; s->rdiv[i] = s->scale; @@ -910,10 +818,8 @@ static av_cold int init(AVFilterContext *ctx) } } else if (!strcmp(ctx->filter->name, "scharr")) { for (i = 0; i < 4; i++) { - if ((1 << i) & s->planes) - s->filter[i] = filter_scharr; - else - s->copy[i] = 1; + s->filter[i] = filter_scharr; + s->copy[i] = !((1 << i) & s->planes); s->size[i] = 3; s->setup[i] = setup_3x3; s->rdiv[i] = s->scale; @@ -921,9 +827,91 @@ static av_cold int init(AVFilterContext *ctx) } } + s->depth = desc->comp[0].depth; + s->max = (1 << s->depth) - 1; + + s->planewidth[1] = s->planewidth[2] = AV_CEIL_RSHIFT(inlink->w, desc->log2_chroma_w); + s->planewidth[0] = s->planewidth[3] = inlink->w; + s->planeheight[1] = s->planeheight[2] = AV_CEIL_RSHIFT(inlink->h, desc->log2_chroma_h); + s->planeheight[0] = s->planeheight[3] = inlink->h; + + s->nb_planes = av_pix_fmt_count_planes(inlink->format); + s->nb_threads = ff_filter_get_nb_threads(ctx); + s->bpc = (s->depth + 7) / 8; + + if (!strcmp(ctx->filter->name, "convolution")) { + if (s->depth > 8) { + for (p = 0; p < s->nb_planes; p++) { + if (s->mode[p] == MATRIX_ROW) + s->filter[p] = filter16_row; + else if (s->mode[p] == MATRIX_COLUMN) + s->filter[p] = filter16_column; + else if (s->size[p] == 3) + s->filter[p] = filter16_3x3; + else if (s->size[p] == 5) + s->filter[p] = filter16_5x5; + else if (s->size[p] == 7) + s->filter[p] = filter16_7x7; + } + } +#if CONFIG_CONVOLUTION_FILTER && ARCH_X86_64 + ff_convolution_init_x86(s); +#endif + } else if (!strcmp(ctx->filter->name, "prewitt")) { + if (s->depth > 8) + for (p = 0; p < s->nb_planes; p++) + s->filter[p] = filter16_prewitt; + } else if (!strcmp(ctx->filter->name, "roberts")) { + if (s->depth > 8) + for (p = 0; p < s->nb_planes; p++) + s->filter[p] = filter16_roberts; + } else if (!strcmp(ctx->filter->name, "sobel")) { + if (s->depth > 8) + for (p = 0; p < s->nb_planes; p++) + s->filter[p] = filter16_sobel; + } else if (!strcmp(ctx->filter->name, "kirsch")) { + if (s->depth > 8) + for (p = 0; p < s->nb_planes; p++) + s->filter[p] = filter16_kirsch; + } else if (!strcmp(ctx->filter->name, "scharr")) { + if (s->depth > 8) + for (p = 0; p < s->nb_planes; p++) + s->filter[p] = filter16_scharr; + } + return 0; } +static int config_input(AVFilterLink *inlink) +{ + AVFilterContext *ctx = inlink->dst; + return param_init(ctx); +} + +static int filter_frame(AVFilterLink *inlink, AVFrame *in) +{ + AVFilterContext *ctx = inlink->dst; + ConvolutionContext *s = ctx->priv; + AVFilterLink *outlink = ctx->outputs[0]; + AVFrame *out; + ThreadData td; + + out = ff_get_video_buffer(outlink, outlink->w, outlink->h); + if (!out) { + av_frame_free(&in); + return AVERROR(ENOMEM); + } + av_frame_copy_props(out, in); + + td.in = in; + td.out = out; + ff_filter_execute(ctx, filter_slice, &td, NULL, + FFMIN3(s->planeheight[1], s->planewidth[1], s->nb_threads)); + + av_frame_free(&in); + return ff_filter_frame(outlink, out); +} + static int process_command(AVFilterContext *ctx, const char *cmd, const char *args, char *res, int res_len, int flags) { @@ -933,7 +921,7 @@ static int process_command(AVFilterContext *ctx, const char *cmd, const char *ar if (ret < 0) return ret; - return init(ctx); + return param_init(ctx); } static const AVFilterPad convolution_inputs[] = { @@ -959,7 +947,6 @@ const AVFilter ff_vf_convolution = { .description = NULL_IF_CONFIG_SMALL("Apply convolution filter."), .priv_size = sizeof(ConvolutionContext), .priv_class = &convolution_class, - .init = init, FILTER_INPUTS(convolution_inputs), FILTER_OUTPUTS(convolution_outputs), FILTER_PIXFMTS_ARRAY(pix_fmts), @@ -986,7 +973,6 @@ const AVFilter ff_vf_prewitt = { .description = NULL_IF_CONFIG_SMALL("Apply prewitt operator."), .priv_size = sizeof(ConvolutionContext), .priv_class = &common_class, - .init = init, FILTER_INPUTS(convolution_inputs), FILTER_OUTPUTS(convolution_outputs), FILTER_PIXFMTS_ARRAY(pix_fmts), @@ -1003,7 +989,6 @@ const AVFilter ff_vf_sobel = { .description = NULL_IF_CONFIG_SMALL("Apply sobel operator."), .priv_size = sizeof(ConvolutionContext), .priv_class = &common_class, - .init = init, FILTER_INPUTS(convolution_inputs), FILTER_OUTPUTS(convolution_outputs), FILTER_PIXFMTS_ARRAY(pix_fmts), @@ -1020,7 +1005,6 @@ const AVFilter ff_vf_roberts = { .description = NULL_IF_CONFIG_SMALL("Apply roberts cross operator."), .priv_size = sizeof(ConvolutionContext), .priv_class = &common_class, - .init = init, FILTER_INPUTS(convolution_inputs), FILTER_OUTPUTS(convolution_outputs), FILTER_PIXFMTS_ARRAY(pix_fmts), @@ -1037,7 +1021,6 @@ const AVFilter ff_vf_kirsch = { .description = NULL_IF_CONFIG_SMALL("Apply kirsch operator."), .priv_size = sizeof(ConvolutionContext), .priv_class = &common_class, - .init = init, FILTER_INPUTS(convolution_inputs), FILTER_OUTPUTS(convolution_outputs), FILTER_PIXFMTS_ARRAY(pix_fmts), @@ -1054,7 +1037,6 @@ const AVFilter ff_vf_scharr = { .description = NULL_IF_CONFIG_SMALL("Apply scharr operator."), .priv_size = sizeof(ConvolutionContext), .priv_class = &common_class, - .init = init, FILTER_INPUTS(convolution_inputs), FILTER_OUTPUTS(convolution_outputs), FILTER_PIXFMTS_ARRAY(pix_fmts),

1 0

avcodec/flac_parser: Consider AV_INPUT_BUFFER_PADDING_SIZE
by Michael Niedermayer 22 Oct '21

22 Oct '21

ffmpeg | branch: master | Michael Niedermayer <michael(a)niedermayer.cc> | Thu Oct 21 13:25:59 2021 +0200| [70024b6b47b9eacfe01e8f92349ca9bf1ccd7d5a] | committer: Michael Niedermayer avcodec/flac_parser: Consider AV_INPUT_BUFFER_PADDING_SIZE Fixes: out if array read Fixes: 40109/clusterfuzz-testcase-minimized-ffmpeg_dem_FLAC_fuzzer-4805686811295744 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Reviewed-by: Mattias Wadman <mattias.wadman(a)gmail.com> Signed-off-by: Michael Niedermayer <michael(a)niedermayer.cc> > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=70024b6b47b9eacfe… --- libavcodec/flac_parser.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/libavcodec/flac_parser.c b/libavcodec/flac_parser.c index 2c550507fc..3b27b152fc 100644 --- a/libavcodec/flac_parser.c +++ b/libavcodec/flac_parser.c @@ -55,6 +55,7 @@ /** largest possible size of flac header */ #define MAX_FRAME_HEADER_SIZE 16 +#define MAX_FRAME_VERIFY_SIZE (MAX_FRAME_HEADER_SIZE + 1) typedef struct FLACHeaderMarker { int offset; /**< byte offset from start of FLACParseContext->buffer */ @@ -99,7 +100,7 @@ static int frame_header_is_valid(AVCodecContext *avctx, const uint8_t *buf, uint8_t subframe_type; // header plus one byte from first subframe - init_get_bits(&gb, buf, MAX_FRAME_HEADER_SIZE * 8 + 8); + init_get_bits(&gb, buf, MAX_FRAME_VERIFY_SIZE * 8); if (ff_flac_decode_frame_header(avctx, &gb, fi, 127)) { return 0; } @@ -196,7 +197,7 @@ static int find_headers_search_validate(FLACParseContext *fpc, int offset) uint8_t *header_buf; int size = 0; header_buf = flac_fifo_read_wrap(fpc, offset, - MAX_FRAME_HEADER_SIZE, + MAX_FRAME_VERIFY_SIZE + AV_INPUT_BUFFER_PADDING_SIZE, &fpc->wrap_buf, &fpc->wrap_buf_allocated_size); if (frame_header_is_valid(fpc->avctx, header_buf, &fi)) {

1 0

avfilter/vf_selectivecolor: no need to use doubles
by Paul B Mahol 22 Oct '21

22 Oct '21

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Fri Oct 22 13:47:16 2021 +0200| [eda2a50c8e9381a357f936a190659bb039e0c038] | committer: Paul B Mahol avfilter/vf_selectivecolor: no need to use doubles > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=eda2a50c8e9381a35… --- libavfilter/vf_selectivecolor.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/libavfilter/vf_selectivecolor.c b/libavfilter/vf_selectivecolor.c index 9ace3ce4a6..ea4b5dfee2 100644 --- a/libavfilter/vf_selectivecolor.c +++ b/libavfilter/vf_selectivecolor.c @@ -223,7 +223,7 @@ static int parse_psfile(AVFilterContext *ctx, const char *fname) int k; for (k = 0; k < FF_ARRAY_ELEMS(s->cmyk_adjust[0]); k++) { READ16(val); - s->cmyk_adjust[i][k] = val / 100.; + s->cmyk_adjust[i][k] = val / 100.f; } ret = register_range(s, i); if (ret < 0) @@ -299,11 +299,11 @@ static const enum AVPixelFormat pix_fmts[] = { static inline int comp_adjust(int scale, float value, float adjust, float k, int correction_method) { const float min = -value; - const float max = 1. - value; - float res = (-1. - adjust) * k - adjust; + const float max = 1.f - value; + float res = (-1.f - adjust) * k - adjust; if (correction_method == CORRECTION_METHOD_RELATIVE) res *= max; - return lrint(av_clipf(res, min, max) * scale); + return lrintf(av_clipf(res, min, max) * scale); } #define DECLARE_SELECTIVE_COLOR_FUNC(nbits) \

1 0

avfilter/vf_bilateral: add slice threading support
by Paul B Mahol 22 Oct '21

22 Oct '21

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Fri Oct 22 11:01:23 2021 +0200| [7d463be590680bcce42aa4039616ac9dfac117b0] | committer: Paul B Mahol avfilter/vf_bilateral: add slice threading support > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=7d463be590680bcce… --- libavfilter/vf_bilateral.c | 461 +++++++++++++++++++++++++++++---------------- 1 file changed, 295 insertions(+), 166 deletions(-) diff --git a/libavfilter/vf_bilateral.c b/libavfilter/vf_bilateral.c index b236efe6c3..41dc38d6f2 100644 --- a/libavfilter/vf_bilateral.c +++ b/libavfilter/vf_bilateral.c @@ -36,6 +36,7 @@ typedef struct BilateralContext { float sigmaR; int planes; + int nb_threads; int nb_planes; int depth; int planewidth[4]; @@ -44,14 +45,14 @@ typedef struct BilateralContext { float alpha; float range_table[65536]; - float *img_out_f; - float *img_temp; - float *map_factor_a; - float *map_factor_b; - float *slice_factor_a; - float *slice_factor_b; - float *line_factor_a; - float *line_factor_b; + float *img_out_f[4]; + float *img_temp[4]; + float *map_factor_a[4]; + float *map_factor_b[4]; + float *slice_factor_a[4]; + float *slice_factor_b[4]; + float *line_factor_a[4]; + float *line_factor_b[4]; } BilateralContext; #define OFFSET(x) offsetof(BilateralContext, x) @@ -102,6 +103,10 @@ static int config_params(AVFilterContext *ctx) return 0; } +typedef struct ThreadData { + AVFrame *in, *out; +} ThreadData; + static int config_input(AVFilterLink *inlink) { AVFilterContext *ctx = inlink->dst; @@ -109,7 +114,6 @@ static int config_input(AVFilterLink *inlink) const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(inlink->format); s->depth = desc->comp[0].depth; - config_params(ctx); s->planewidth[1] = s->planewidth[2] = AV_CEIL_RSHIFT(inlink->w, desc->log2_chroma_w); @@ -118,133 +122,169 @@ static int config_input(AVFilterLink *inlink) s->planeheight[0] = s->planeheight[3] = inlink->h; s->nb_planes = av_pix_fmt_count_planes(inlink->format); - - s->img_out_f = av_calloc(inlink->w * inlink->h, sizeof(float)); - s->img_temp = av_calloc(inlink->w * inlink->h, sizeof(float)); - s->map_factor_a = av_calloc(inlink->w * inlink->h, sizeof(float)); - s->map_factor_b = av_calloc(inlink->w * inlink->h, sizeof(float)); - s->slice_factor_a = av_calloc(inlink->w, sizeof(float)); - s->slice_factor_b = av_calloc(inlink->w, sizeof(float)); - s->line_factor_a = av_calloc(inlink->w, sizeof(float)); - s->line_factor_b = av_calloc(inlink->w, sizeof(float)); - - if (!s->img_out_f || - !s->img_temp || - !s->map_factor_a || - !s->map_factor_b || - !s->slice_factor_a || - !s->slice_factor_a || - !s->line_factor_a || - !s->line_factor_a) - return AVERROR(ENOMEM); + s->nb_threads = ff_filter_get_nb_threads(ctx); + + for (int p = 0; p < s->nb_planes; p++) { + const int w = s->planewidth[p]; + const int h = s->planeheight[p]; + + s->img_out_f[p] = av_calloc(w * h, sizeof(float)); + s->img_temp[p] = av_calloc(w * h, sizeof(float)); + s->map_factor_a[p] = av_calloc(w * h, sizeof(float)); + s->map_factor_b[p] = av_calloc(w * h, sizeof(float)); + s->slice_factor_a[p] = av_calloc(w, sizeof(float)); + s->slice_factor_b[p] = av_calloc(w, sizeof(float)); + s->line_factor_a[p] = av_calloc(w, sizeof(float)); + s->line_factor_b[p] = av_calloc(w, sizeof(float)); + + if (!s->img_out_f[p] || + !s->img_temp[p] || + !s->map_factor_a[p] || + !s->map_factor_b[p] || + !s->slice_factor_a[p] || + !s->slice_factor_a[p] || + !s->line_factor_a[p] || + !s->line_factor_a[p]) + return AVERROR(ENOMEM); + } return 0; } -#define BILATERAL(type, name) \ -static void bilateral_##name(BilateralContext *s, const uint8_t *ssrc, uint8_t *ddst, \ - int width, int height, int src_linesize, int dst_linesize) \ -{ \ - type *dst = (type *)ddst; \ - const type *src = (const type *)ssrc; \ - float *img_out_f = s->img_out_f, *img_temp = s->img_temp; \ - float *map_factor_a = s->map_factor_a, *map_factor_b = s->map_factor_b; \ - float *slice_factor_a = s->slice_factor_a, *slice_factor_b = s->slice_factor_b; \ - float *line_factor_a = s->line_factor_a, *line_factor_b = s->line_factor_b; \ - const float *range_table = s->range_table; \ - const float alpha = s->alpha; \ - float ypr, ycr, *ycy, *ypy, *xcy, fp, fc; \ - const float inv_alpha_ = 1.f - alpha; \ - float *ycf, *ypf, *xcf, *in_factor; \ - const type *tcy, *tpy; \ - int h1; \ - \ - for (int y = 0; y < height; y++) { \ - float *temp_factor_x, *temp_x = &img_temp[y * width]; \ - const type *in_x = &src[y * src_linesize]; \ - const type *texture_x = &src[y * src_linesize]; \ - type tpr; \ - \ - *temp_x++ = ypr = *in_x++; \ - tpr = *texture_x++; \ - \ - temp_factor_x = &map_factor_a[y * width]; \ - *temp_factor_x++ = fp = 1; \ - \ - for (int x = 1; x < width; x++) { \ - float alpha_; \ - int range_dist; \ - type tcr = *texture_x++; \ - type dr = abs(tcr - tpr); \ - \ - range_dist = dr; \ - alpha_ = range_table[range_dist]; \ - *temp_x++ = ycr = inv_alpha_*(*in_x++) + alpha_*ypr; \ - tpr = tcr; \ - ypr = ycr; \ - *temp_factor_x++ = fc = inv_alpha_ + alpha_ * fp; \ - fp = fc; \ - } \ - --temp_x; *temp_x = ((*temp_x) + (*--in_x)); \ - tpr = *--texture_x; \ - ypr = *in_x; \ - \ - --temp_factor_x; *temp_factor_x = ((*temp_factor_x) + 1); \ - fp = 1; \ - \ - for (int x = width - 2; x >= 0; x--) { \ - type tcr = *--texture_x; \ - type dr = abs(tcr - tpr); \ - int range_dist = dr; \ - float alpha_ = range_table[range_dist]; \ - \ - ycr = inv_alpha_ * (*--in_x) + alpha_ * ypr; \ - --temp_x; *temp_x = ((*temp_x) + ycr); \ - tpr = tcr; \ - ypr = ycr; \ - \ - fc = inv_alpha_ + alpha_*fp; \ - --temp_factor_x; \ - *temp_factor_x = ((*temp_factor_x) + fc); \ - fp = fc; \ - } \ - } \ - memcpy(img_out_f, img_temp, sizeof(float) * width); \ - \ - in_factor = map_factor_a; \ - memcpy(map_factor_b, in_factor, sizeof(float) * width); \ - for (int y = 1; y < height; y++) { \ - tpy = &src[(y - 1) * src_linesize]; \ - tcy = &src[y * src_linesize]; \ - xcy = &img_temp[y * width]; \ - ypy = &img_out_f[(y - 1) * width]; \ - ycy = &img_out_f[y * width]; \ - \ - xcf = &in_factor[y * width]; \ - ypf = &map_factor_b[(y - 1) * width]; \ - ycf = &map_factor_b[y * width]; \ - for (int x = 0; x < width; x++) { \ - type dr = abs((*tcy++) - (*tpy++)); \ - int range_dist = dr; \ - float alpha_ = range_table[range_dist]; \ - \ - *ycy++ = inv_alpha_*(*xcy++) + alpha_*(*ypy++); \ - *ycf++ = inv_alpha_*(*xcf++) + alpha_*(*ypf++); \ - } \ - } \ - h1 = height - 1; \ - ycf = line_factor_a; \ - ypf = line_factor_b; \ - memcpy(ypf, &in_factor[h1 * width], sizeof(float) * width); \ - for (int x = 0; x < width; x++) \ - map_factor_b[h1 * width + x] = (map_factor_b[h1 * width + x] + ypf[x]); \ - \ - ycy = slice_factor_a; \ - ypy = slice_factor_b; \ - memcpy(ypy, &img_temp[h1 * width], sizeof(float) * width); \ - for (int x = 0, k = 0; x < width; x++) { \ - int idx = h1 * width + x; \ - img_out_f[idx] = (img_out_f[idx] + ypy[k++]) / map_factor_b[h1 * width + x]; \ +#define BILATERAL_H(type, name) \ +static void bilateralh_##name(BilateralContext *s, AVFrame *out, AVFrame *in, \ + int jobnr, int nb_jobs, int plane) \ +{ \ + const int width = s->planewidth[plane]; \ + const int height = s->planeheight[plane]; \ + const int slice_start = (height * jobnr) / nb_jobs; \ + const int slice_end = (height * (jobnr+1)) / nb_jobs; \ + const int src_linesize = in->linesize[plane] / sizeof(type); \ + const type *src = (const type *)in->data[plane]; \ + float *img_temp = s->img_temp[plane]; \ + float *map_factor_a = s->map_factor_a[plane]; \ + const float *const range_table = s->range_table; \ + const float alpha = s->alpha; \ + float ypr, ycr, fp, fc; \ + const float inv_alpha_ = 1.f - alpha; \ + \ + for (int y = slice_start; y < slice_end; y++) { \ + float *temp_factor_x, *temp_x = &img_temp[y * width]; \ + const type *in_x = &src[y * src_linesize]; \ + const type *texture_x = &src[y * src_linesize]; \ + type tpr; \ + \ + *temp_x++ = ypr = *in_x++; \ + tpr = *texture_x++; \ + \ + temp_factor_x = &map_factor_a[y * width]; \ + *temp_factor_x++ = fp = 1; \ + \ + for (int x = 1; x < width; x++) { \ + float alpha_; \ + int range_dist; \ + type tcr = *texture_x++; \ + type dr = abs(tcr - tpr); \ + \ + range_dist = dr; \ + alpha_ = range_table[range_dist]; \ + *temp_x++ = ycr = inv_alpha_*(*in_x++) + alpha_*ypr; \ + tpr = tcr; \ + ypr = ycr; \ + *temp_factor_x++ = fc = inv_alpha_ + alpha_ * fp; \ + fp = fc; \ + } \ + --temp_x; *temp_x = ((*temp_x) + (*--in_x)); \ + tpr = *--texture_x; \ + ypr = *in_x; \ + \ + --temp_factor_x; *temp_factor_x = ((*temp_factor_x) + 1); \ + fp = 1; \ + \ + for (int x = width - 2; x >= 0; x--) { \ + type tcr = *--texture_x; \ + type dr = abs(tcr - tpr); \ + int range_dist = dr; \ + float alpha_ = range_table[range_dist]; \ + \ + ycr = inv_alpha_ * (*--in_x) + alpha_ * ypr; \ + --temp_x; *temp_x = ((*temp_x) + ycr); \ + tpr = tcr; \ + ypr = ycr; \ + \ + fc = inv_alpha_ + alpha_*fp; \ + --temp_factor_x; \ + *temp_factor_x = ((*temp_factor_x) + fc); \ + fp = fc; \ + } \ + } \ +} + +BILATERAL_H(uint8_t, byte) +BILATERAL_H(uint16_t, word) + +#define BILATERAL_V(type, name) \ +static void bilateralv_##name(BilateralContext *s, AVFrame *out, AVFrame *in, \ + int jobnr, int nb_jobs, int plane) \ +{ \ + const int width = s->planewidth[plane]; \ + const int height = s->planeheight[plane]; \ + const int slice_start = (width * jobnr) / nb_jobs; \ + const int slice_end = (width * (jobnr+1)) / nb_jobs; \ + const int src_linesize = in->linesize[plane] / sizeof(type); \ + const type *src = (const type *)in->data[plane] + slice_start; \ + float *img_out_f = s->img_out_f[plane] + slice_start; \ + float *img_temp = s->img_temp[plane] + slice_start; \ + float *map_factor_a = s->map_factor_a[plane] + slice_start; \ + float *map_factor_b = s->map_factor_b[plane] + slice_start; \ + float *slice_factor_a = s->slice_factor_a[plane] + slice_start; \ + float *slice_factor_b = s->slice_factor_b[plane] + slice_start; \ + float *line_factor_a = s->line_factor_a[plane] + slice_start; \ + float *line_factor_b = s->line_factor_b[plane] + slice_start; \ + const float *const range_table = s->range_table; \ + const float alpha = s->alpha; \ + float *ycy, *ypy, *xcy; \ + const float inv_alpha_ = 1.f - alpha; \ + float *ycf, *ypf, *xcf, *in_factor; \ + const type *tcy, *tpy; \ + int h1; \ + \ + memcpy(img_out_f, img_temp, sizeof(float) * (slice_end - slice_start)); \ + \ + in_factor = map_factor_a; \ + memcpy(map_factor_b, in_factor, sizeof(float) * (slice_end - slice_start)); \ + for (int y = 1; y < height; y++) { \ + tpy = &src[(y - 1) * src_linesize]; \ + tcy = &src[y * src_linesize]; \ + xcy = &img_temp[y * width]; \ + ypy = &img_out_f[(y - 1) * width]; \ + ycy = &img_out_f[y * width]; \ + \ + xcf = &in_factor[y * width]; \ + ypf = &map_factor_b[(y - 1) * width]; \ + ycf = &map_factor_b[y * width]; \ + for (int x = 0; x < slice_end - slice_start; x++) { \ + type dr = abs((*tcy++) - (*tpy++)); \ + int range_dist = dr; \ + float alpha_ = range_table[range_dist]; \ + \ + *ycy++ = inv_alpha_*(*xcy++) + alpha_*(*ypy++); \ + *ycf++ = inv_alpha_*(*xcf++) + alpha_*(*ypf++); \ + } \ + } \ + h1 = height - 1; \ + ycf = line_factor_a; \ + ypf = line_factor_b; \ + memcpy(ypf, &in_factor[h1 * width], sizeof(float) * (slice_end - slice_start)); \ + for (int x = 0, k = 0; x < slice_end - slice_start; x++) \ + map_factor_b[h1 * width + x] = (map_factor_b[h1 * width + x] + ypf[k++]); \ + \ + ycy = slice_factor_a; \ + ypy = slice_factor_b; \ + memcpy(ypy, &img_temp[h1 * width], sizeof(float) * (slice_end - slice_start)); \ + for (int x = 0, k = 0; x < slice_end - slice_start; x++) { \ + int idx = h1 * width + x; \ + img_out_f[idx] = (img_out_f[idx] + ypy[k++]) / map_factor_b[h1 * width + x]; \ } \ \ for (int y = h1 - 1; y >= 0; y--) { \ @@ -262,7 +302,7 @@ static void bilateral_##name(BilateralContext *s, const uint8_t *ssrc, uint8_t * ycf_ = ycf; \ ypf_ = ypf; \ factor_ = &map_factor_b[y * width]; \ - for (int x = 0; x < width; x++) { \ + for (int x = 0; x < slice_end - slice_start; x++) { \ type dr = abs((*tcy++) - (*tpy++)); \ int range_dist = dr; \ float alpha_ = range_table[range_dist]; \ @@ -281,20 +321,119 @@ static void bilateral_##name(BilateralContext *s, const uint8_t *ssrc, uint8_t * ypy = ycy; \ ypf = ycf; \ } \ - \ - for (int i = 0; i < height; i++) \ - for (int j = 0; j < width; j++) \ - dst[j + i * dst_linesize] = lrintf(img_out_f[i * width + j]); \ } -BILATERAL(uint8_t, byte) -BILATERAL(uint16_t, word) +BILATERAL_V(uint8_t, byte) +BILATERAL_V(uint16_t, word) + +#define BILATERAL_O(type, name) \ +static void bilateralo_##name(BilateralContext *s, AVFrame *out, AVFrame *in, \ + int jobnr, int nb_jobs, int plane) \ +{ \ + const int width = s->planewidth[plane]; \ + const int height = s->planeheight[plane]; \ + const int slice_start = (height * jobnr) / nb_jobs; \ + const int slice_end = (height * (jobnr+1)) / nb_jobs; \ + const int dst_linesize = out->linesize[plane] / sizeof(type); \ + \ + for (int i = slice_start; i < slice_end; i++) { \ + type *dst = (type *)out->data[plane] + i * dst_linesize; \ + const float *const img_out_f = s->img_out_f[plane] + i * width; \ + for (int j = 0; j < width; j++) \ + dst[j] = lrintf(img_out_f[j]); \ + } \ +} + +BILATERAL_O(uint8_t, byte) +BILATERAL_O(uint16_t, word) + +static int bilateralh_planes(AVFilterContext *ctx, void *arg, + int jobnr, int nb_jobs) +{ + BilateralContext *s = ctx->priv; + ThreadData *td = arg; + AVFrame *out = td->out; + AVFrame *in = td->in; + + for (int plane = 0; plane < s->nb_planes; plane++) { + if (!(s->planes & (1 << plane))) + continue; + + if (s->depth <= 8) + bilateralh_byte(s, out, in, jobnr, nb_jobs, plane); + else + bilateralh_word(s, out, in, jobnr, nb_jobs, plane); + } + + return 0; +} + +static int bilateralv_planes(AVFilterContext *ctx, void *arg, + int jobnr, int nb_jobs) +{ + BilateralContext *s = ctx->priv; + ThreadData *td = arg; + AVFrame *out = td->out; + AVFrame *in = td->in; + + for (int plane = 0; plane < s->nb_planes; plane++) { + if (!(s->planes & (1 << plane))) + continue; + + if (s->depth <= 8) + bilateralv_byte(s, out, in, jobnr, nb_jobs, plane); + else + bilateralv_word(s, out, in, jobnr, nb_jobs, plane); + } + + return 0; +} + +static int bilateralo_planes(AVFilterContext *ctx, void *arg, + int jobnr, int nb_jobs) +{ + BilateralContext *s = ctx->priv; + ThreadData *td = arg; + AVFrame *out = td->out; + AVFrame *in = td->in; + + for (int plane = 0; plane < s->nb_planes; plane++) { + if (!(s->planes & (1 << plane))) { + if (out != in) { + const int height = s->planeheight[plane]; + const int slice_start = (height * jobnr) / nb_jobs; + const int slice_end = (height * (jobnr+1)) / nb_jobs; + const int width = s->planewidth[plane]; + const int linesize = in->linesize[plane]; + const int dst_linesize = out->linesize[plane]; + const uint8_t *src = in->data[plane]; + uint8_t *dst = out->data[plane]; + + av_image_copy_plane(dst + slice_start * dst_linesize, + dst_linesize, + src + slice_start * linesize, + linesize, + width * ((s->depth + 7) / 8), + slice_end - slice_start); + } + continue; + } + + if (s->depth <= 8) + bilateralo_byte(s, out, in, jobnr, nb_jobs, plane); + else + bilateralo_word(s, out, in, jobnr, nb_jobs, plane); + } + + return 0; +} static int filter_frame(AVFilterLink *inlink, AVFrame *in) { AVFilterContext *ctx = inlink->dst; BilateralContext *s = ctx->priv; AVFilterLink *outlink = ctx->outputs[0]; + ThreadData td; AVFrame *out; if (av_frame_is_writable(in)) { @@ -308,24 +447,11 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in) av_frame_copy_props(out, in); } - for (int plane = 0; plane < s->nb_planes; plane++) { - if (!(s->planes & (1 << plane))) { - if (out != in) - av_image_copy_plane(out->data[plane], out->linesize[plane], - in->data[plane], in->linesize[plane], - s->planewidth[plane] * ((s->depth + 7) / 8), s->planeheight[plane]); - continue; - } - - if (s->depth <= 8) - bilateral_byte(s, in->data[plane], out->data[plane], - s->planewidth[plane], s->planeheight[plane], - in->linesize[plane], out->linesize[plane]); - else - bilateral_word(s, in->data[plane], out->data[plane], - s->planewidth[plane], s->planeheight[plane], - in->linesize[plane] / 2, out->linesize[plane] / 2); - } + td.in = in; + td.out = out; + ff_filter_execute(ctx, bilateralh_planes, &td, NULL, s->nb_threads); + ff_filter_execute(ctx, bilateralv_planes, &td, NULL, s->nb_threads); + ff_filter_execute(ctx, bilateralo_planes, &td, NULL, s->nb_threads); if (out != in) av_frame_free(&in); @@ -336,14 +462,16 @@ static av_cold void uninit(AVFilterContext *ctx) { BilateralContext *s = ctx->priv; - av_freep(&s->img_out_f); - av_freep(&s->img_temp); - av_freep(&s->map_factor_a); - av_freep(&s->map_factor_b); - av_freep(&s->slice_factor_a); - av_freep(&s->slice_factor_b); - av_freep(&s->line_factor_a); - av_freep(&s->line_factor_b); + for (int p = 0; p < s->nb_planes; p++) { + av_freep(&s->img_out_f[p]); + av_freep(&s->img_temp[p]); + av_freep(&s->map_factor_a[p]); + av_freep(&s->map_factor_b[p]); + av_freep(&s->slice_factor_a[p]); + av_freep(&s->slice_factor_b[p]); + av_freep(&s->line_factor_a[p]); + av_freep(&s->line_factor_b[p]); + } } static int process_command(AVFilterContext *ctx, @@ -386,6 +514,7 @@ const AVFilter ff_vf_bilateral = { FILTER_INPUTS(bilateral_inputs), FILTER_OUTPUTS(bilateral_outputs), FILTER_PIXFMTS_ARRAY(pix_fmts), - .flags = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC, + .flags = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC | + AVFILTER_FLAG_SLICE_THREADS, .process_command = process_command, };

1 0

avfilter/vf_bilateral: add direct aka writable frame support
by Paul B Mahol 22 Oct '21

22 Oct '21

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Fri Oct 22 09:34:51 2021 +0200| [444cf3fca8dab2da9b95cc36920667f76d0f796c] | committer: Paul B Mahol avfilter/vf_bilateral: add direct aka writable frame support > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=444cf3fca8dab2da9… --- libavfilter/vf_bilateral.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) diff --git a/libavfilter/vf_bilateral.c b/libavfilter/vf_bilateral.c index 8fe341f70d..b236efe6c3 100644 --- a/libavfilter/vf_bilateral.c +++ b/libavfilter/vf_bilateral.c @@ -297,18 +297,23 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in) AVFilterLink *outlink = ctx->outputs[0]; AVFrame *out; - out = ff_get_video_buffer(outlink, outlink->w, outlink->h); - if (!out) { - av_frame_free(&in); - return AVERROR(ENOMEM); + if (av_frame_is_writable(in)) { + out = in; + } else { + out = ff_get_video_buffer(outlink, outlink->w, outlink->h); + if (!out) { + av_frame_free(&in); + return AVERROR(ENOMEM); + } + av_frame_copy_props(out, in); } - av_frame_copy_props(out, in); for (int plane = 0; plane < s->nb_planes; plane++) { if (!(s->planes & (1 << plane))) { - av_image_copy_plane(out->data[plane], out->linesize[plane], - in->data[plane], in->linesize[plane], - s->planewidth[plane] * ((s->depth + 7) / 8), s->planeheight[plane]); + if (out != in) + av_image_copy_plane(out->data[plane], out->linesize[plane], + in->data[plane], in->linesize[plane], + s->planewidth[plane] * ((s->depth + 7) / 8), s->planeheight[plane]); continue; } @@ -322,7 +327,8 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in) in->linesize[plane] / 2, out->linesize[plane] / 2); } - av_frame_free(&in); + if (out != in) + av_frame_free(&in); return ff_filter_frame(outlink, out); }

1 0

avfilter/vf_bilateral: properly round float result
by Paul B Mahol 21 Oct '21

21 Oct '21

ffmpeg | branch: master | Paul B Mahol <onemda(a)gmail.com> | Thu Oct 21 21:59:14 2021 +0200| [cf0881bcfc88b4afb62a4647b0e09809e9612a5b] | committer: Paul B Mahol avfilter/vf_bilateral: properly round float result > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=cf0881bcfc88b4afb… --- libavfilter/vf_bilateral.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavfilter/vf_bilateral.c b/libavfilter/vf_bilateral.c index aecde64860..8fe341f70d 100644 --- a/libavfilter/vf_bilateral.c +++ b/libavfilter/vf_bilateral.c @@ -284,7 +284,7 @@ static void bilateral_##name(BilateralContext *s, const uint8_t *ssrc, uint8_t * \ for (int i = 0; i < height; i++) \ for (int j = 0; j < width; j++) \ - dst[j + i * dst_linesize] = img_out_f[i * width + j]; \ + dst[j + i * dst_linesize] = lrintf(img_out_f[i * width + j]); \ } BILATERAL(uint8_t, byte)

1 0