mirror of https://git.ffmpeg.org/ffmpeg.git
avfilter/vf_nlmeans: refactor line processing in preparation for x86 SIMD assembly
This commit is contained in:
parent
38c86e7a02
commit
eaca36d495
|
@ -38,11 +38,6 @@
|
|||
#include "vf_nlmeans.h"
|
||||
#include "video.h"
|
||||
|
||||
struct weighted_avg {
|
||||
float total_weight;
|
||||
float sum;
|
||||
};
|
||||
|
||||
typedef struct NLMeansContext {
|
||||
const AVClass *class;
|
||||
int nb_planes;
|
||||
|
@ -329,6 +324,58 @@ struct thread_data {
|
|||
int p;
|
||||
};
|
||||
|
||||
static void compute_weights_line_c(const uint32_t *const iia,
|
||||
const uint32_t *const iib,
|
||||
const uint32_t *const iid,
|
||||
const uint32_t *const iie,
|
||||
const uint8_t *const src,
|
||||
struct weighted_avg *wa,
|
||||
const float *const weight_lut,
|
||||
int max_meaningful_diff,
|
||||
int startx, int endx)
|
||||
{
|
||||
for (int x = startx; x < endx; x++) {
|
||||
/*
|
||||
* M is a discrete map where every entry contains the sum of all the entries
|
||||
* in the rectangle from the top-left origin of M to its coordinate. In the
|
||||
* following schema, "i" contains the sum of the whole map:
|
||||
*
|
||||
* M = +----------+-----------------+----+
|
||||
* | | | |
|
||||
* | | | |
|
||||
* | a| b| c|
|
||||
* +----------+-----------------+----+
|
||||
* | | | |
|
||||
* | | | |
|
||||
* | | X | |
|
||||
* | | | |
|
||||
* | d| e| f|
|
||||
* +----------+-----------------+----+
|
||||
* | | | |
|
||||
* | g| h| i|
|
||||
* +----------+-----------------+----+
|
||||
*
|
||||
* The sum of the X box can be calculated with:
|
||||
* X = e-d-b+a
|
||||
*
|
||||
* See https://en.wikipedia.org/wiki/Summed_area_table
|
||||
*
|
||||
* The compute*_ssd functions compute the integral image M where every entry
|
||||
* contains the sum of the squared difference of every corresponding pixels of
|
||||
* two input planes of the same size as M.
|
||||
*/
|
||||
const uint32_t a = iia[x];
|
||||
const uint32_t b = iib[x];
|
||||
const uint32_t d = iid[x];
|
||||
const uint32_t e = iie[x];
|
||||
const uint32_t patch_diff_sq = FFMIN(e - d - b + a, max_meaningful_diff);
|
||||
const float weight = weight_lut[patch_diff_sq]; // exp(-patch_diff_sq * s->pdiff_scale)
|
||||
|
||||
wa[x].total_weight += weight;
|
||||
wa[x].sum += weight * src[x];
|
||||
}
|
||||
}
|
||||
|
||||
static int nlmeans_slice(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
|
||||
{
|
||||
NLMeansContext *s = ctx->priv;
|
||||
|
@ -346,50 +393,19 @@ static int nlmeans_slice(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs
|
|||
const int dist_d = dist_b * s->ii_lz_32;
|
||||
const int dist_e = dist_d + dist_b;
|
||||
const float *const weight_lut = s->weight_lut;
|
||||
NLMeansDSPContext *dsp = &s->dsp;
|
||||
|
||||
for (int y = starty; y < endy; y++) {
|
||||
const uint8_t *src = td->src + y*src_linesize;
|
||||
const uint8_t *const src = td->src + y*src_linesize;
|
||||
struct weighted_avg *wa = s->wa + y*s->wa_linesize;
|
||||
for (int x = td->startx; x < td->endx; x++) {
|
||||
/*
|
||||
* M is a discrete map where every entry contains the sum of all the entries
|
||||
* in the rectangle from the top-left origin of M to its coordinate. In the
|
||||
* following schema, "i" contains the sum of the whole map:
|
||||
*
|
||||
* M = +----------+-----------------+----+
|
||||
* | | | |
|
||||
* | | | |
|
||||
* | a| b| c|
|
||||
* +----------+-----------------+----+
|
||||
* | | | |
|
||||
* | | | |
|
||||
* | | X | |
|
||||
* | | | |
|
||||
* | d| e| f|
|
||||
* +----------+-----------------+----+
|
||||
* | | | |
|
||||
* | g| h| i|
|
||||
* +----------+-----------------+----+
|
||||
*
|
||||
* The sum of the X box can be calculated with:
|
||||
* X = e-d-b+a
|
||||
*
|
||||
* See https://en.wikipedia.org/wiki/Summed_area_table
|
||||
*
|
||||
* The compute*_ssd functions compute the integral image M where every entry
|
||||
* contains the sum of the squared difference of every corresponding pixels of
|
||||
* two input planes of the same size as M.
|
||||
*/
|
||||
const uint32_t a = ii[x];
|
||||
const uint32_t b = ii[x + dist_b];
|
||||
const uint32_t d = ii[x + dist_d];
|
||||
const uint32_t e = ii[x + dist_e];
|
||||
const uint32_t patch_diff_sq = FFMIN(e - d - b + a, max_meaningful_diff);
|
||||
const float weight = weight_lut[patch_diff_sq]; // exp(-patch_diff_sq * s->pdiff_scale)
|
||||
const uint32_t *const iia = ii;
|
||||
const uint32_t *const iib = ii + dist_b;
|
||||
const uint32_t *const iid = ii + dist_d;
|
||||
const uint32_t *const iie = ii + dist_e;
|
||||
|
||||
wa[x].total_weight += weight;
|
||||
wa[x].sum += weight * src[x];
|
||||
}
|
||||
dsp->compute_weights_line(iia, iib, iid, iie, src, wa,
|
||||
weight_lut, max_meaningful_diff,
|
||||
td->startx, td->endx);
|
||||
ii += s->ii_lz_32;
|
||||
}
|
||||
return 0;
|
||||
|
@ -493,6 +509,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *in)
|
|||
void ff_nlmeans_init(NLMeansDSPContext *dsp)
|
||||
{
|
||||
dsp->compute_safe_ssd_integral_image = compute_safe_ssd_integral_image_c;
|
||||
dsp->compute_weights_line = compute_weights_line_c;
|
||||
|
||||
if (ARCH_AARCH64)
|
||||
ff_nlmeans_init_aarch64(dsp);
|
||||
|
|
|
@ -22,11 +22,25 @@
|
|||
#include <stddef.h>
|
||||
#include <stdint.h>
|
||||
|
||||
struct weighted_avg {
|
||||
float total_weight;
|
||||
float sum;
|
||||
};
|
||||
|
||||
typedef struct NLMeansDSPContext {
|
||||
void (*compute_safe_ssd_integral_image)(uint32_t *dst, ptrdiff_t dst_linesize_32,
|
||||
const uint8_t *s1, ptrdiff_t linesize1,
|
||||
const uint8_t *s2, ptrdiff_t linesize2,
|
||||
int w, int h);
|
||||
void (*compute_weights_line)(const uint32_t *const iia,
|
||||
const uint32_t *const iib,
|
||||
const uint32_t *const iid,
|
||||
const uint32_t *const iie,
|
||||
const uint8_t *const src,
|
||||
struct weighted_avg *wa,
|
||||
const float *const weight_lut,
|
||||
int max_meaningful_diff,
|
||||
int startx, int endx);
|
||||
} NLMeansDSPContext;
|
||||
|
||||
void ff_nlmeans_init(NLMeansDSPContext *dsp);
|
||||
|
|
Loading…
Reference in New Issue