[FFmpeg-devel] [GSoC'22] Added Chromakey CUDA filter
Timo Rothenpieler
timo at rothenpieler.org
Sun Jul 3 19:07:34 EEST 2022
On 30.06.2022 02:01, mohamed Elhadidy wrote:
> GSoC'22 GPU accelerated video filters
> Added CUDA chromakeyfilter
> libavfilter/vf_chromakey_cuda.cu:the CUDA kernel for the filter
> libavfilter/vf_chromakey_cuda.c: the C side that calls the kernel and gets user input
> libavfilter/allfilters.c: added the filter to it
> libavfilter/Makefile: added the filter to it
> cuda/cuda_runtime.h: added two math CUDA functions that are used in the filter
> ---
> compat/cuda/cuda_runtime.h | 2 +
> libavfilter/Makefile | 2 +
> libavfilter/allfilters.c | 1 +
> libavfilter/vf_chromakey_cuda.c | 520 +++++++++++++++++++++++++++++++
> libavfilter/vf_chromakey_cuda.cu | 248 +++++++++++++++
> 5 files changed, 773 insertions(+)
> create mode 100644 libavfilter/vf_chromakey_cuda.c
> create mode 100644 libavfilter/vf_chromakey_cuda.cu
>
> diff --git a/compat/cuda/cuda_runtime.h b/compat/cuda/cuda_runtime.h
> index 30cd085e48..5837c1ad37 100644
> --- a/compat/cuda/cuda_runtime.h
> +++ b/compat/cuda/cuda_runtime.h
> @@ -181,7 +181,9 @@ static inline __device__ double trunc(double a) { return __builtin_trunc(a); }
> static inline __device__ float fabsf(float a) { return __builtin_fabsf(a); }
> static inline __device__ float fabs(float a) { return __builtin_fabsf(a); }
> static inline __device__ double fabs(double a) { return __builtin_fabs(a); }
> +static inline __device__ float sqrtf(float a) { return __builtin_sqrtf(a); }
>
> +static inline __device__ float __saturatef(float a) { return __nvvm_saturate_f(a); }
> static inline __device__ float __sinf(float a) { return __nvvm_sin_approx_f(a); }
> static inline __device__ float __cosf(float a) { return __nvvm_cos_approx_f(a); }
> static inline __device__ float __expf(float a) { return __nvvm_ex2_approx_f(a * (float)__builtin_log2(__builtin_exp(1))); }
> diff --git a/libavfilter/Makefile b/libavfilter/Makefile
> index 22b0a0ca15..8aee10fc76 100644
> --- a/libavfilter/Makefile
> +++ b/libavfilter/Makefile
> @@ -210,6 +210,8 @@ OBJS-$(CONFIG_CAS_FILTER) += vf_cas.o
> OBJS-$(CONFIG_CHROMABER_VULKAN_FILTER) += vf_chromaber_vulkan.o vulkan.o vulkan_filter.o
> OBJS-$(CONFIG_CHROMAHOLD_FILTER) += vf_chromakey.o
> OBJS-$(CONFIG_CHROMAKEY_FILTER) += vf_chromakey.o
> +OBJS-$(CONFIG_CHROMAKEY_CUDA_FILTER) += vf_chromakey_cuda.o vf_chromakey_cuda.ptx.o framesync.o
This doesn't use framesync at all, does it?
Only the overlay filter does.
> OBJS-$(CONFIG_CHROMANR_FILTER) += vf_chromanr.o
> OBJS-$(CONFIG_CHROMASHIFT_FILTER) += vf_chromashift.o
> OBJS-$(CONFIG_CIESCOPE_FILTER) += vf_ciescope.o
> diff --git a/libavfilter/allfilters.c b/libavfilter/allfilters.c
> index ec70feef11..da1a96b23c 100644
> --- a/libavfilter/allfilters.c
> +++ b/libavfilter/allfilters.c
> @@ -195,6 +195,7 @@ extern const AVFilter ff_vf_cas;
> extern const AVFilter ff_vf_chromaber_vulkan;
> extern const AVFilter ff_vf_chromahold;
> extern const AVFilter ff_vf_chromakey;
> +extern const AVFilter ff_vf_chromakey_cuda;
> extern const AVFilter ff_vf_chromanr;
> extern const AVFilter ff_vf_chromashift;
> extern const AVFilter ff_vf_ciescope;
> diff --git a/libavfilter/vf_chromakey_cuda.c b/libavfilter/vf_chromakey_cuda.c
> new file mode 100644
> index 0000000000..822488d436
> --- /dev/null
> +++ b/libavfilter/vf_chromakey_cuda.c
> @@ -0,0 +1,520 @@
> +/*
> +* Copyright (c) 2017, NVIDIA CORPORATION. All rights reserved.
This probably should be changed, there's basically none of the original
code left.
> +* Permission is hereby granted, free of charge, to any person obtaining a
> +* copy of this software and associated documentation files (the "Software"),
> +* to deal in the Software without restriction, including without limitation
> +* the rights to use, copy, modify, merge, publish, distribute, sublicense,
> +* and/or sell copies of the Software, and to permit persons to whom the
> +* Software is furnished to do so, subject to the following conditions:
> +*
> +* The above copyright notice and this permission notice shall be included in
> +* all copies or substantial portions of the Software.
> +*
> +* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> +* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> +* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
> +* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> +* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
> +* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
> +* DEALINGS IN THE SOFTWARE.
> +*/
> +
> +#include <float.h>
> +#include <stdio.h>
> +#include <string.h>
> +
> +#include "libavutil/avstring.h"
> +#include "libavutil/common.h"
> +#include "libavutil/hwcontext.h"
> +#include "libavutil/hwcontext_cuda_internal.h"
> +#include "libavutil/cuda_check.h"
> +#include "libavutil/internal.h"
> +#include "libavutil/opt.h"
> +#include "libavutil/pixdesc.h"
> +
> +#include "avfilter.h"
> +#include "formats.h"
> +#include "internal.h"
> +#include "video.h"
> +
> +#include "cuda/load_helper.h"
> +
> +
> +
> +#define FIXNUM(x) lrint((x) * (1 << 10))
> +#define RGB_TO_U(rgb) (((- FIXNUM(0.16874) * rgb[0] - FIXNUM(0.33126) * rgb[1] + FIXNUM(0.50000) * rgb[2] + (1 << 9) - 1) >> 10) + 128)
> +#define RGB_TO_V(rgb) ((( FIXNUM(0.50000) * rgb[0] - FIXNUM(0.41869) * rgb[1] - FIXNUM(0.08131) * rgb[2] + (1 << 9) - 1) >> 10) + 128)
> +
> +
> +static const enum AVPixelFormat supported_formats[] = {
> + AV_PIX_FMT_YUV420P,
> + AV_PIX_FMT_NV12,
> + AV_PIX_FMT_YUV444P,
> + AV_PIX_FMT_YUVA420P
> +
spurious empty line
> +};
Does the filter actually support yuv444p?
Looking at the kernels, it seems to be hardcoded for 4:2:0 subsampling.
Should be easy to adapt though, see comment in Kernel below.
> +#define DIV_UP(a, b) ( ((a) + (b) - 1) / (b) )
> +#define BLOCKX 32
> +#define BLOCKY 16
> +
> +#define CHECK_CU(x) FF_CUDA_CHECK_DL(ctx, s->hwctx->internal->cuda_dl, x)
> +
> +
> +typedef struct ChromakeyCUDAContext {
> + const AVClass *class;
> +
> + AVCUDADeviceContext *hwctx;
> +
> + enum AVPixelFormat in_fmt, out_fmt;
> + const AVPixFmtDescriptor *in_desc, *out_desc;
> + int in_planes, out_planes;
> + int in_plane_depths[4];
> + int in_plane_channels[4];
> +
> +
> + uint8_t chromakey_rgba[4];
> + uint16_t chromakey_uv[2];
> + int is_yuv;
> + float similarity;
> + float blend;
> +
> +
> +
Superfluous empty lines here and above in the struct.
> + AVBufferRef *frames_ctx;
> + AVFrame *frame;
> +
> + AVFrame *tmp_frame;
> +
> + char *w_expr; ///< width expression string
> + char *h_expr; ///< height expression string
Comments can probably go, given no other elements here have them.
> + CUcontext cu_ctx;
> + CUmodule cu_module;
> + CUfunction cu_func;
> + CUfunction cu_func_uv;
> + CUstream cu_stream;
> +} ChromakeyCUDAContext;
> +
> +static av_cold int cudachromakey_init(AVFilterContext *ctx)
> +{
> + ChromakeyCUDAContext *s = ctx->priv;
> +
> + s->frame = av_frame_alloc();
> + if (!s->frame)
> + return AVERROR(ENOMEM);
> +
> + s->tmp_frame = av_frame_alloc();
> + if (!s->tmp_frame)
> + return AVERROR(ENOMEM);
> +
> + return 0;
> +}
> +
> +static av_cold void cudachromakey_uninit(AVFilterContext *ctx)
> +{
> + ChromakeyCUDAContext *s = ctx->priv;
> +
> + if (s->hwctx && s->cu_module) {
> + CudaFunctions *cu = s->hwctx->internal->cuda_dl;
> + CUcontext context;
> +
> + CHECK_CU(cu->cuCtxPushCurrent(s->hwctx->cuda_ctx));
> + CHECK_CU(cu->cuModuleUnload(s->cu_module));
> + s->cu_module = NULL;
> + CHECK_CU(cu->cuCtxPopCurrent(&context));
> + }
> +
> + av_frame_free(&s->frame);
> + av_buffer_unref(&s->frames_ctx);
> + av_frame_free(&s->tmp_frame);
> +}
> +
> +static av_cold int init_hwframe_ctx(ChromakeyCUDAContext *s, AVBufferRef *device_ctx, int width, int height)
> +{
> + AVBufferRef *out_ref = NULL;
> + AVHWFramesContext *out_ctx;
> + int ret;
> +
> + out_ref = av_hwframe_ctx_alloc(device_ctx);
> + if (!out_ref)
> + return AVERROR(ENOMEM);
> + out_ctx = (AVHWFramesContext*)out_ref->data;
> +
> + out_ctx->format = AV_PIX_FMT_CUDA;
> + out_ctx->sw_format = s->out_fmt;
> + out_ctx->width = width;
> + out_ctx->height = height;
> +
> + ret = av_hwframe_ctx_init(out_ref);
> + if (ret < 0)
> + goto fail;
> +
> + av_frame_unref(s->frame);
> + ret = av_hwframe_get_buffer(out_ref, s->frame, 0);
> + if (ret < 0)
> + goto fail;
> +
> + av_buffer_unref(&s->frames_ctx);
> + s->frames_ctx = out_ref;
> +
> + return 0;
> +fail:
> + av_buffer_unref(&out_ref);
> + return ret;
> +}
> +
> +static int format_is_supported(enum AVPixelFormat fmt)
> +{
> + int i;
> +
> + for (i = 0; i < FF_ARRAY_ELEMS(supported_formats); i++)
> + if (supported_formats[i] == fmt)
> + return 1;
> + return 0;
> +}
> +
> +static av_cold void set_format_info(AVFilterContext *ctx, enum AVPixelFormat in_format, enum AVPixelFormat out_format)
> +{
> + ChromakeyCUDAContext *s = ctx->priv;
> + int i, p, d;
> +
> + s->in_fmt = in_format;
> + s->out_fmt = out_format;
> +
> + s->in_desc = av_pix_fmt_desc_get(s->in_fmt);
> + s->out_desc = av_pix_fmt_desc_get(s->out_fmt);
> + s->in_planes = av_pix_fmt_count_planes(s->in_fmt);
> + s->out_planes = av_pix_fmt_count_planes(s->out_fmt);
> +
> + // find maximum step of each component of each plane
> + // For our subset of formats, this should accurately tell us how many channels CUDA needs
> + // i.e. 1 for Y plane, 2 for UV plane of NV12, 4 for single plane of RGB0 formats
> +
> + for (i = 0; i < s->in_desc->nb_components; i++) {
> + d = (s->in_desc->comp[i].depth + 7) / 8;
> + p = s->in_desc->comp[i].plane;
> + s->in_plane_channels[p] = FFMAX(s->in_plane_channels[p], s->in_desc->comp[i].step / d);
> +
> + s->in_plane_depths[p] = s->in_desc->comp[i].depth;
> + }
> +}
> +
> +static av_cold int init_processing_chain(AVFilterContext *ctx, int width, int height)
> +{
> + ChromakeyCUDAContext *s = ctx->priv;
> + AVHWFramesContext *in_frames_ctx;
> + int ret;
> +
> + /* check that we have a hw context */
> + if (!ctx->inputs[0]->hw_frames_ctx) {
> + av_log(ctx, AV_LOG_ERROR, "No hw context provided on input\n");
> + return AVERROR(EINVAL);
> + }
> + in_frames_ctx = (AVHWFramesContext*)ctx->inputs[0]->hw_frames_ctx->data;
> +
> + if (!format_is_supported(in_frames_ctx->sw_format)) {
> + av_log(ctx, AV_LOG_ERROR, "Unsupported format: %s\n", av_get_pix_fmt_name(in_frames_ctx->sw_format));
> + return AVERROR(ENOSYS);
> + }
> +
> + set_format_info(ctx, in_frames_ctx->sw_format,AV_PIX_FMT_YUVA420P );
> +
> + ret = init_hwframe_ctx(s, in_frames_ctx->device_ref, width, height);
> + if (ret < 0)
> + return ret;
> +
> + ctx->outputs[0]->hw_frames_ctx = av_buffer_ref(s->frames_ctx);
> + if (!ctx->outputs[0]->hw_frames_ctx)
> + return AVERROR(ENOMEM);
> +
> + return 0;
> +}
> +
> +static av_cold int cudachromakey_load_functions(AVFilterContext *ctx)
> +{
> + ChromakeyCUDAContext *s = ctx->priv;
> + CUcontext context, cuda_ctx = s->hwctx->cuda_ctx;
> + CudaFunctions *cu = s->hwctx->internal->cuda_dl;
> + int ret;
> +
> + extern const unsigned char ff_vf_chromakey_cuda_ptx_data[];
> + extern const unsigned int ff_vf_chromakey_cuda_ptx_len;
> +
> + ret = CHECK_CU(cu->cuCtxPushCurrent(cuda_ctx));
> + if (ret < 0)
> + return ret;
> +
> + ret = ff_cuda_load_module(ctx, s->hwctx, &s->cu_module,
> + ff_vf_chromakey_cuda_ptx_data, ff_vf_chromakey_cuda_ptx_len);
> + if (ret < 0)
> + goto fail;
> +
> + ret = CHECK_CU(cu->cuModuleGetFunction(&s->cu_func, s->cu_module, "Process_uchar"));
> + if (ret < 0) {
> + av_log(ctx, AV_LOG_FATAL, "Failed loading Process_uchar\n");
> + goto fail;
> + }
> +
> + ret = CHECK_CU(cu->cuModuleGetFunction(&s->cu_func_uv, s->cu_module, "Process_uchar2"));
> + if (ret < 0) {
> + av_log(ctx, AV_LOG_FATAL, "Failed loading Process_uchar2\n");
> + goto fail;
> + }
> +
> +fail:
> + CHECK_CU(cu->cuCtxPopCurrent(&context));
> +
> + return ret;
> +}
> +
> +static av_cold int cudachromakey_config_props(AVFilterLink *outlink)
> +{
> + AVFilterContext *ctx = outlink->src;
> + AVFilterLink *inlink = outlink->src->inputs[0];
> + ChromakeyCUDAContext *s = ctx->priv;
> + AVHWFramesContext *frames_ctx = (AVHWFramesContext*)inlink->hw_frames_ctx->data;
> + AVCUDADeviceContext *device_hwctx = frames_ctx->device_ctx->hwctx;
> + int ret;
> +
> + s->hwctx = device_hwctx;
> + s->cu_stream = s->hwctx->stream;
> +
> + ret = init_processing_chain(ctx, inlink->w, inlink->h);
> + if (ret < 0)
> + return ret;
> +
> + outlink->sample_aspect_ratio = inlink->sample_aspect_ratio;
> +
> + ret = cudachromakey_load_functions(ctx);
> + if (ret < 0)
> + return ret;
> +
> + return 0;
> +}
> +
> +static int call_cuda_kernel(AVFilterContext *ctx, CUfunction func,
> + CUtexObject src_tex[3], AVFrame *out_frame,
> + int width, int height, int pitch,
> + int width_uv, int height_uv, int pitch_uv,
> + float u_key,float v_key, float similarity,
> + float blend )
> +{
> + ChromakeyCUDAContext *s = ctx->priv;
> + CudaFunctions *cu = s->hwctx->internal->cuda_dl;
> + int ret;
> +
> + CUdeviceptr dst_devptr[4] = {
> + (CUdeviceptr)out_frame->data[0], (CUdeviceptr)out_frame->data[1],
> + (CUdeviceptr)out_frame->data[2], (CUdeviceptr)out_frame->data[3]
> +
> + };
> +
> + void *args_uchar[] = {
> + &src_tex[0], &src_tex[1], &src_tex[2],
> + &dst_devptr[0], &dst_devptr[1], &dst_devptr[2],&dst_devptr[3],
> + &width, &height, &pitch,
> + &width_uv, &height_uv, &pitch_uv,&u_key,&v_key,&similarity,&blend
> + };
> +
> + ret = CHECK_CU(cu->cuLaunchKernel(func,
> + DIV_UP(width, BLOCKX), DIV_UP(height, BLOCKY), 1,
> + BLOCKX, BLOCKY, 1, 0, s->cu_stream, args_uchar, NULL));
> + if (ret < 0)
> + return ret;
> +
> + // This ruins performance, but catches errors that happened in the Kernel immediately
> + // Remove for performance/production test
> + ret = CHECK_CU(cu->cuStreamSynchronize(s->cu_stream));
> +
> + return ret;
> +}
> +
> +static int cudachromakey_process_internal(AVFilterContext *ctx,
> + AVFrame *out, AVFrame *in)
> +{
> + ChromakeyCUDAContext *s = ctx->priv;
> + CudaFunctions *cu = s->hwctx->internal->cuda_dl;
> + CUcontext context, cuda_ctx = s->hwctx->cuda_ctx;
> + float u_key,v_key;
> + int i, ret;
> +
> + CUtexObject tex[3] = { 0, 0, 0 };
> + if(s->is_yuv){
> + u_key=s->chromakey_rgba[1];
> + v_key=s->chromakey_rgba[2];
> +
> + }else{
> + u_key = RGB_TO_U(s->chromakey_rgba);
> + v_key = RGB_TO_V(s->chromakey_rgba);
> + }
> + ret = CHECK_CU(cu->cuCtxPushCurrent(cuda_ctx));
> + if (ret < 0)
> + return ret;
> +
> + for (i = 0; i < s->in_planes; i++) {
> + CUDA_TEXTURE_DESC tex_desc = {
> + .filterMode = CU_TR_FILTER_MODE_LINEAR,
> + .flags = 0, // CU_TRSF_READ_AS_INTEGER to get raw ints instead of normalized floats from tex2D
> + };
> +
> + CUDA_RESOURCE_DESC res_desc = {
> + .resType = CU_RESOURCE_TYPE_PITCH2D,
> + .res.pitch2D.format = CU_AD_FORMAT_UNSIGNED_INT8,
> + .res.pitch2D.numChannels = s->in_plane_channels[i],
> + .res.pitch2D.pitchInBytes = in->linesize[i],
> + .res.pitch2D.devPtr = (CUdeviceptr)in->data[i],
> + };
> +
> + if (i == 1 || i == 2) {
> + res_desc.res.pitch2D.width = AV_CEIL_RSHIFT(in->width, s->in_desc->log2_chroma_w);
> + res_desc.res.pitch2D.height = AV_CEIL_RSHIFT(in->height, s->in_desc->log2_chroma_h);
> + } else {
> + res_desc.res.pitch2D.width = in->width;
> + res_desc.res.pitch2D.height = in->height;
> + }
> +
> + ret = CHECK_CU(cu->cuTexObjectCreate(&tex[i], &res_desc, &tex_desc, NULL));
> + if (ret < 0)
> + goto exit;
> + }
> +
> + ret = call_cuda_kernel(ctx, (s->in_plane_channels[1] > 1) ? s->cu_func_uv : s->cu_func,
> + tex, out,
> + out->width, out->height, out->linesize[0],
> + AV_CEIL_RSHIFT(out->width, s->out_desc->log2_chroma_w),
> + AV_CEIL_RSHIFT(out->height, s->out_desc->log2_chroma_h),
> + out->linesize[1],
> + u_key,v_key,s->similarity,s->blend );
extra whitespace before )
> + if (ret < 0)
> + goto exit;
> +
> +exit:
> + for (i = 0; i < s->in_planes; i++)
> + if (tex[i])
> + CHECK_CU(cu->cuTexObjectDestroy(tex[i]));
> +
> + CHECK_CU(cu->cuCtxPopCurrent(&context));
> +
> + return ret;
> +}
> +
> +static int cudachromakey_process(AVFilterContext *ctx, AVFrame *out, AVFrame *in)
> +{
> + ChromakeyCUDAContext *s = ctx->priv;
> + AVFrame *src = in;
> + int ret;
> +
> + ret = cudachromakey_process_internal(ctx, s->frame, src);
> + if (ret < 0)
> + return ret;
> +
> + src = s->frame;
> + ret = av_hwframe_get_buffer(src->hw_frames_ctx, s->tmp_frame, 0);
> + if (ret < 0)
> + return ret;
> +
> + av_frame_move_ref(out, s->frame);
> + av_frame_move_ref(s->frame, s->tmp_frame);
> +
> + ret = av_frame_copy_props(out, in);
> + if (ret < 0)
> + return ret;
> +
> + return 0;
> +}
> +
> +static int cudachromakey_filter_frame(AVFilterLink *link, AVFrame *in)
> +{
> + AVFilterContext *ctx = link->dst;
> + ChromakeyCUDAContext *s = ctx->priv;
the '=' got out of alignment with the others
> + AVFilterLink *outlink = ctx->outputs[0];
> + CudaFunctions *cu = s->hwctx->internal->cuda_dl;
> +
> + AVFrame *out = NULL;
> + CUcontext context;
> + int ret = 0;
> +
> + out = av_frame_alloc();
> + if (!out) {
> + ret = AVERROR(ENOMEM);
> + goto fail;
> + }
> +
> + ret = CHECK_CU(cu->cuCtxPushCurrent(s->hwctx->cuda_ctx));
> + if (ret < 0)
> + goto fail;
> +
> + ret = cudachromakey_process(ctx, out, in);
> +
> + CHECK_CU(cu->cuCtxPopCurrent(&context));
> + if (ret < 0)
> + goto fail;
> +
> + av_frame_free(&in);
> + return ff_filter_frame(outlink, out);
> +fail:
> + av_frame_free(&in);
> + av_frame_free(&out);
> + return ret;
> +}
> +
> +static AVFrame *cudachromakey_get_video_buffer(AVFilterLink *inlink, int w, int h)
> +{
> + return ff_default_get_video_buffer(inlink, w, h);
> +}
> +
> +#define OFFSET(x) offsetof(ChromakeyCUDAContext, x)
> +#define FLAGS (AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM)
> +static const AVOption options[] = {
> + { "color", "set the chromakey key color", OFFSET(chromakey_rgba), AV_OPT_TYPE_COLOR, { .str = "black" }, 0, 0, FLAGS },
> + { "similarity", "set the chromakey similarity value", OFFSET(similarity), AV_OPT_TYPE_FLOAT, { .dbl = 0.01 }, 0.01, 1.0, FLAGS },
> + { "blend", "set the chromakey key blend value", OFFSET(blend), AV_OPT_TYPE_FLOAT, { .dbl = 0.0 }, 0.0, 1.0, FLAGS },
> + { "yuv", "color parameter is in yuv instead of rgb", OFFSET(is_yuv), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, FLAGS },
> + { "w", "Output video width", OFFSET(w_expr), AV_OPT_TYPE_STRING, { .str = "iw" }, .flags = FLAGS },
> + { "h", "Output video height", OFFSET(h_expr), AV_OPT_TYPE_STRING, { .str = "ih" }, .flags = FLAGS },
> + { NULL },
> +};
> +
> +static const AVClass cudachromakey_class = {
> + .class_name = "cudachromakey",
> + .item_name = av_default_item_name,
> + .option = options,
> + .version = LIBAVUTIL_VERSION_INT,
> +};
> +
> +static const AVFilterPad cudachromakey_inputs[] = {
> + {
> + .name = "default",
> + .type = AVMEDIA_TYPE_VIDEO,
> + .filter_frame = cudachromakey_filter_frame,
> + .get_buffer.video = cudachromakey_get_video_buffer,
> + },
> +};
> +
> +static const AVFilterPad cudachromakey_outputs[] = {
> + {
> + .name = "default",
> + .type = AVMEDIA_TYPE_VIDEO,
> + .config_props = cudachromakey_config_props,
> + },
> +};
> +
> +const AVFilter ff_vf_chromakey_cuda = {
> + .name = "chromakey_cuda",
> + .description = NULL_IF_CONFIG_SMALL("GPU accelerated chromakey filter"),
> +
> + .init = cudachromakey_init,
> + .uninit = cudachromakey_uninit,
> +
> + .priv_size = sizeof(ChromakeyCUDAContext),
> + .priv_class = &cudachromakey_class,
> +
> + FILTER_INPUTS(cudachromakey_inputs),
> + FILTER_OUTPUTS(cudachromakey_outputs),
> +
> + FILTER_SINGLE_PIXFMT(AV_PIX_FMT_CUDA),
> +
> + .flags_internal = FF_FILTER_FLAG_HWFRAME_AWARE,
> +};
> diff --git a/libavfilter/vf_chromakey_cuda.cu b/libavfilter/vf_chromakey_cuda.cu
> new file mode 100644
> index 0000000000..fef118a0f8
> --- /dev/null
> +++ b/libavfilter/vf_chromakey_cuda.cu
> @@ -0,0 +1,248 @@
> +/*
> + * Copyright (c) 2022, Chromakey
> + *
> + * Permission is hereby granted, free of charge, to any person obtaining a
> + * copy of this software and associated documentation files (the "Software"),
> + * to deal in the Software without restriction, including without limitation
> + * the rights to use, copy, modify, merge, publish, distribute, sublicense,
> + * and/or sell copies of the Software, and to permit persons to whom the
> + * Software is furnished to do so, subject to the following conditions:
> + *
> + * The above copyright notice and this permission notice shall be included in
> + * all copies or substantial portions of the Software.
> + *
> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
> + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
> + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
> + * DEALINGS IN THE SOFTWARE.
> + */
> +
> +#include "cuda/vector_helpers.cuh"
> +
> +extern "C"
> +{
> +
> +
> +/**
> + * @brief function contains the main logic of chroma keying, and changes the alpahc channel with the suitable value
> + *
> + * @param src_tex texture U or texture UV , decided based on the passed is_uchar2 flag
> + * @param src_tex_V texture V , used only if is_uchar2 flag is false
> + * @param dst_A alpha channel destination
> + * @param width_uv width of uv channels
> + * @param height_uv height of uv channels
> + * @param width width of alpha channel
> + * @param height height of alpha channel
> + * @param pitch pitch of alpha channel
> + * @param x current x coordinate of pixel
> + * @param y current y coordinate of pixel
> + * @param chromakey_uv uv values for chroma keying
> + * @param similarity similarity of keying
> + * @param blend blend of keying
> + * @param is_uchar2 true means that sent texture is uchar2 for the UV channels, false means U and V are sent seperatley
> + * @param resize_ratio the ratio of alpha channel to UV channels
> + * @return void
> + */
> +__device__ static inline void change_alpha_channel(
> + cudaTextureObject_t src_tex, cudaTextureObject_t src_tex_V,
> + uchar *dst_A, int width_uv,
> + int height_uv, int width,
> + int height, int pitch,
> + int x, int y,
Can you rearrange those, so that corresponding values share a line?
As in:
int width_uv, int height_uv,
int width, int height, int pitch,
And so on? Having them intermixed like this makes this quite hard to
read imo. Same goes for the other two functions.
> + float2 chromakey_uv, float similarity,
> + float blend, bool is_uchar2,
> + uchar resize_ratio)
> +{
> +
Nit: we don't typically have empty first lines in functions
> + uchar window_size = 3;
> + int start_r = x - window_size / 2;
> + int start_c = y - window_size / 2;
> +
> + uchar counter = 0;
> + float diff = 0.0f;
> + float du, dv;
> +
> + // loop over the eight neighbourhood of the current pixel(x,y)
> + for (uchar i = 0; i < window_size; i++)
> + {
> + for (uchar j = 0; j < window_size; j++)
> + {
> + int r = start_r + i;
> + int c = start_c + j;
> + bool check_flag = (r >= 0 && r < width_uv && c >= 0 && c < height_uv);
> +
> + if (!check_flag)
> + continue;
> +
> + float u_value, v_value;
> + if (is_uchar2){
You should be able to just check if src_tex_V is valid, saving you one
of the already quite numerous parameters.
> + float2 temp_uv = tex2D<float2>(src_tex, r, c);
> + u_value = temp_uv.x;
> + v_value = temp_uv.y;
> + }else{
> + u_value = tex2D<float>(src_tex, r, c);
> + v_value = tex2D<float>(src_tex_V, r, c);
> + }
> +
> + du = (u_value * 255.0f) - chromakey_uv.x;
> + dv = (v_value * 255.0f) - chromakey_uv.y;
> + diff += sqrtf((du * du + dv * dv) / (255.0f * 255.0f * 2.f));
> + counter++;
> + }
> + }
> +
> + if (counter > 0){
> + diff = diff / counter;
> + }else{
> + diff /= 9.0f;
> + }
> +
> + uchar alpha_value;
One too many whitespaces.
> + if(blend>0.0001f){
> + alpha_value=__saturatef((diff - similarity) / blend)*255;
> + }else{
> + alpha_value=(diff < similarity ? 0 : 1)*255;
> + }
No {} for single-line if/else statements.
> + //write the value in the aloha channel with regarding the ratio of (alpha_size : uv_size)
Typo, "alpha"
> + for (uchar k = 0; k < resize_ratio; k++)
> + {
> + for (uchar l = 0; l < resize_ratio; l++)
> + {
> + int x_resize = x * resize_ratio + k;
> + int y_resize = y * resize_ratio + l;
> + int a_channel_resize = y_resize * pitch + x_resize;
> + if (y_resize >= height || x_resize >= width)
> + continue;
> + dst_A[a_channel_resize] = alpha_value;
> + }
> + }
> +
> +
Two more spurious empty lines.
> +}
> +
> +
> +/**
> + * @brief Function to process yuv420p video , and pass it to change_alpha_channel function, the output video is yuva420p
> + *
> + * @param src_tex_Y Y channel of input video frame
> + * @param src_tex_U U channel of input video frame
> + * @param src_tex_V V channel of input video frame
> + * @param dst_Y Y channel of output video frame
> + * @param dst_U U channel of output video frame
> + * @param dst_V V channel of output video frame
> + * @param dst_A A channel of output video frame
> + * @param width width of Y channel for input,video frame
> + * @param height height of Y channel for input,video frame
> + * @param pitch pitch(line size) of Y channel for input,video frame
> + * @param width_uv width of UV for input,output video frame
> + * @param height_uv height of UV for input,output video frame
> + * @param pitch_uv pitch(line size) of UV for input,output video frame
> + * @param u_key u part of chromakey (user input)
> + * @param v_key v part of chromakey (user input)
> + * @param similarity similairty user want (user input)
> + * @param blend blend user want (user input)
> + * @return void
> + */
You don't need to document all of these for all the functions, given
they're all pretty much identical.
Describing them on the shared function is more than plenty.
> +__global__ void Process_uchar(
> + cudaTextureObject_t src_tex_Y, cudaTextureObject_t src_tex_U,
> + cudaTextureObject_t src_tex_V,uchar *dst_Y,
> + uchar *dst_U, uchar *dst_V,
> + uchar *dst_A,int width,
> + int height, int pitch,
> + int width_uv, int height_uv,
> + int pitch_uv,float u_key,
> + float v_key, float similarity,
> + float blend)
> +{
> +
> + uchar resize_ratio = 2;
If you manage to derive this from the actual information given, this
function could otherwise support yuv444p as-is.
Like, if it's 444, width_uv will be equal to width.
For 420, width_uv will be only half of it.
> + int x = blockIdx.x * blockDim.x + threadIdx.x;
> + int y = blockIdx.y * blockDim.y + threadIdx.y;
> +
> + if (y >= height || x >= width)
> + return;
> + dst_Y[y * pitch + x] = tex2D<float>(src_tex_Y, x, y)*255;
> + if (y >= height_uv || x >= width_uv)
> + return;
> +
> + int uv_index = y * pitch_uv + x;
> + dst_U[uv_index]=tex2D<float>(src_tex_U,x,y)*255;
> + dst_V[uv_index]=tex2D<float>(src_tex_V,x,y)*255;
> +
> +
> +
> + change_alpha_channel(src_tex_U,src_tex_V,
> + dst_A,width_uv,height_uv,
> + width,height,
> + pitch,x,
> + y,make_float2(u_key,v_key),
> + similarity,blend,
> + false,resize_ratio);
> +
> +}
> +
> +
> +
> +/**
> + * @brief Function to process nv12 video , and pass it to change_alpha_channel function, the output video is yuva420p
> + *
> + * @param src_tex_Y Y channel of input video frame
> + * @param src_tex_UV UV channels of input video frame
> + * @param unused1 unused parameter
> + * @param dst_Y Y channel of output video frame
> + * @param dst_U U channel of output video frame
> + * @param dst_V V channel of output video frame
> + * @param dst_A A channel of output video frame
> + * @param width width of Y channel for input,video frame
> + * @param height height of Y channel for input,video frame
> + * @param pitch pitch(line size) of Y channel for input,video frame
> + * @param width_uv width of UV for input,output video frame
> + * @param height_uv height of UV for input,output video frame
> + * @param pitch_uv pitch(line size) of UV for input, output video frame
> + * @param u_key u part of chromakey (user input)
> + * @param v_key v part of chromakey (user input)
> + * @param similarity similairty user want (user input)
> + * @param blend blend user want (user input)
> + * @return void
> + */
> +__global__ void Process_uchar2(
> + cudaTextureObject_t src_tex_Y, cudaTextureObject_t src_tex_UV,
> + cudaTextureObject_t unused1,uchar *dst_Y,
> + uchar *dst_U, uchar *dst_V,
> + uchar *dst_A,int width,
> + int height, int pitch,
> + int width_uv, int height_uv,
> + int pitch_uv,float u_key,
> + float v_key, float similarity,
> + float blend)
> +{
> +
> + uchar resize_ratio = 2; // alphachannel_size : uv_channel_size ratio
> + int x = blockIdx.x * blockDim.x + threadIdx.x; // x coordinate of current pixel
> + int y = blockIdx.y * blockDim.y + threadIdx.y; // y coordinate of current pixel
> +
> + if (y >= height || x >= width)
> + return;
> + dst_Y[y * pitch + x] = tex2D<float>(src_tex_Y, x, y)*255;
> +
> + if (y >= height_uv || x >= width_uv)
> + return;
> + int uv_index= y * pitch_uv + x;
> + float2 uv_temp=tex2D<float2>(src_tex_UV,x,y);
> + dst_U[uv_index]=uv_temp.x*255;
> + dst_V[uv_index]=uv_temp.y*255;
> +
> + change_alpha_channel(src_tex_UV, unused1,
> + dst_A, width_uv,
> + height_uv, width,
> + height, pitch,
> + x, y,
> + make_float2(u_key,v_key), similarity,
> + blend, true,
> + resize_ratio);
> +}
> +}
Looks good otherwise, and works as expected for all I could test, thank you!
Make sure to mark a re-send of the fixed patch as v2 and send it as
reply to this mail, so it shows up in the same thread.
More information about the ffmpeg-devel
mailing list