FFmpeg
dnn_filter_common.c
Go to the documentation of this file.
1 /*
2  * This file is part of FFmpeg.
3  *
4  * FFmpeg is free software; you can redistribute it and/or
5  * modify it under the terms of the GNU Lesser General Public
6  * License as published by the Free Software Foundation; either
7  * version 2.1 of the License, or (at your option) any later version.
8  *
9  * FFmpeg is distributed in the hope that it will be useful,
10  * but WITHOUT ANY WARRANTY; without even the implied warranty of
11  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
12  * Lesser General Public License for more details.
13  *
14  * You should have received a copy of the GNU Lesser General Public
15  * License along with FFmpeg; if not, write to the Free Software
16  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
17  */
18 
19 #include "dnn_filter_common.h"
20 #include "libavutil/avstring.h"
21 
22 #define MAX_SUPPORTED_OUTPUTS_NB 4
23 
24 static char **separate_output_names(const char *expr, const char *val_sep, int *separated_nb)
25 {
26  char *val, **parsed_vals = NULL;
27  int val_num = 0;
28  if (!expr || !val_sep || !separated_nb) {
29  return NULL;
30  }
31 
32  parsed_vals = av_calloc(MAX_SUPPORTED_OUTPUTS_NB, sizeof(*parsed_vals));
33  if (!parsed_vals) {
34  return NULL;
35  }
36 
37  do {
38  val = av_get_token(&expr, val_sep);
39  if(val) {
40  parsed_vals[val_num] = val;
41  val_num++;
42  }
43  if (*expr) {
44  expr++;
45  }
46  } while(*expr);
47 
48  parsed_vals[val_num] = NULL;
49  *separated_nb = val_num;
50 
51  return parsed_vals;
52 }
53 
55 {
56  DNNBackendType backend = ctx->backend_type;
57 
58  if (!ctx->model_filename) {
59  av_log(filter_ctx, AV_LOG_ERROR, "model file for network is not specified\n");
60  return AVERROR(EINVAL);
61  }
62 
63  if (backend == DNN_TH) {
64  if (ctx->model_inputname)
65  av_log(filter_ctx, AV_LOG_WARNING, "LibTorch backend do not require inputname, "\
66  "inputname will be ignored.\n");
67  if (ctx->model_outputnames)
68  av_log(filter_ctx, AV_LOG_WARNING, "LibTorch backend do not require outputname(s), "\
69  "all outputname(s) will be ignored.\n");
70  ctx->nb_outputs = 1;
71  } else if (backend == DNN_TF) {
72  if (!ctx->model_inputname) {
73  av_log(filter_ctx, AV_LOG_ERROR, "input name of the model network is not specified\n");
74  return AVERROR(EINVAL);
75  }
76  ctx->model_outputnames = separate_output_names(ctx->model_outputnames_string, "&", &ctx->nb_outputs);
77  if (!ctx->model_outputnames) {
78  av_log(filter_ctx, AV_LOG_ERROR, "could not parse model output names\n");
79  return AVERROR(EINVAL);
80  }
81  }
82 
83  ctx->dnn_module = ff_get_dnn_module(ctx->backend_type, filter_ctx);
84  if (!ctx->dnn_module) {
85  av_log(filter_ctx, AV_LOG_ERROR, "could not create DNN module for requested backend\n");
86  return AVERROR(ENOMEM);
87  }
88  if (!ctx->dnn_module->load_model) {
89  av_log(filter_ctx, AV_LOG_ERROR, "load_model for network is not specified\n");
90  return AVERROR(EINVAL);
91  }
92 
93  ctx->model = (ctx->dnn_module->load_model)(ctx->model_filename, func_type, ctx->backend_options, filter_ctx);
94  if (!ctx->model) {
95  av_log(filter_ctx, AV_LOG_ERROR, "could not load DNN model\n");
96  return AVERROR(EINVAL);
97  }
98 
99  return 0;
100 }
101 
103 {
104  ctx->model->frame_pre_proc = pre_proc;
105  ctx->model->frame_post_proc = post_proc;
106  return 0;
107 }
108 
110 {
111  ctx->model->detect_post_proc = post_proc;
112  return 0;
113 }
114 
116 {
117  ctx->model->classify_post_proc = post_proc;
118  return 0;
119 }
120 
122 {
123  return ctx->model->get_input(ctx->model->model, input, ctx->model_inputname);
124 }
125 
126 int ff_dnn_get_output(DnnContext *ctx, int input_width, int input_height, int *output_width, int *output_height)
127 {
128  char * output_name = ctx->model_outputnames && ctx->backend_type != DNN_TH ?
129  ctx->model_outputnames[0] : NULL;
130  return ctx->model->get_output(ctx->model->model, ctx->model_inputname, input_width, input_height,
131  (const char *)output_name, output_width, output_height);
132 }
133 
134 int ff_dnn_execute_model(DnnContext *ctx, AVFrame *in_frame, AVFrame *out_frame)
135 {
136  DNNExecBaseParams exec_params = {
137  .input_name = ctx->model_inputname,
138  .output_names = (const char **)ctx->model_outputnames,
139  .nb_output = ctx->nb_outputs,
140  .in_frame = in_frame,
141  .out_frame = out_frame,
142  };
143  return (ctx->dnn_module->execute_model)(ctx->model, &exec_params);
144 }
145 
146 int ff_dnn_execute_model_classification(DnnContext *ctx, AVFrame *in_frame, AVFrame *out_frame, const char *target)
147 {
148  DNNExecClassificationParams class_params = {
149  {
150  .input_name = ctx->model_inputname,
151  .output_names = (const char **)ctx->model_outputnames,
152  .nb_output = ctx->nb_outputs,
153  .in_frame = in_frame,
154  .out_frame = out_frame,
155  },
156  .target = target,
157  };
158  return (ctx->dnn_module->execute_model)(ctx->model, &class_params.base);
159 }
160 
162 {
163  return (ctx->dnn_module->get_result)(ctx->model, in_frame, out_frame);
164 }
165 
167 {
168  return (ctx->dnn_module->flush)(ctx->model);
169 }
170 
172 {
173  if (ctx->dnn_module) {
174  (ctx->dnn_module->free_model)(&ctx->model);
175  }
176  if (ctx->model_outputnames) {
177  for (int i = 0; i < ctx->nb_outputs; i++)
178  av_free(ctx->model_outputnames[i]);
179 
180  av_freep(&ctx->model_outputnames);
181  }
182 }
AV_LOG_WARNING
#define AV_LOG_WARNING
Something somehow does not look correct.
Definition: log.h:186
MAX_SUPPORTED_OUTPUTS_NB
#define MAX_SUPPORTED_OUTPUTS_NB
Definition: dnn_filter_common.c:22
AVERROR
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later. That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another. Frame references ownership and permissions
DNNFunctionType
DNNFunctionType
Definition: dnn_interface.h:52
AVFrame
This structure describes decoded (raw) audio or video data.
Definition: frame.h:375
DetectPostProc
int(* DetectPostProc)(AVFrame *frame, DNNData *output, uint32_t nb, AVFilterContext *filter_ctx)
Definition: dnn_interface.h:90
separate_output_names
static char ** separate_output_names(const char *expr, const char *val_sep, int *separated_nb)
Definition: dnn_filter_common.c:24
DNNExecBaseParams::input_name
const char * input_name
Definition: dnn_interface.h:77
dnn_filter_common.h
DnnContext
Definition: dnn_filter_common.h:29
filter_ctx
static FilteringContext * filter_ctx
Definition: transcode.c:51
val
static double val(void *priv, double ch)
Definition: aeval.c:78
DNN_TF
@ DNN_TF
Definition: dnn_interface.h:35
AV_LOG_ERROR
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:180
DNNExecClassificationParams
Definition: dnn_interface.h:84
ff_dnn_set_detect_post_proc
int ff_dnn_set_detect_post_proc(DnnContext *ctx, DetectPostProc post_proc)
Definition: dnn_filter_common.c:109
DNNData
Definition: dnn_interface.h:65
ff_dnn_get_result
DNNAsyncStatusType ff_dnn_get_result(DnnContext *ctx, AVFrame **in_frame, AVFrame **out_frame)
Definition: dnn_filter_common.c:161
ctx
AVFormatContext * ctx
Definition: movenc.c:48
ff_dnn_get_input
int ff_dnn_get_input(DnnContext *ctx, DNNData *input)
Definition: dnn_filter_common.c:121
DNNExecClassificationParams::base
DNNExecBaseParams base
Definition: dnn_interface.h:85
NULL
#define NULL
Definition: coverity.c:32
ff_dnn_execute_model_classification
int ff_dnn_execute_model_classification(DnnContext *ctx, AVFrame *in_frame, AVFrame *out_frame, const char *target)
Definition: dnn_filter_common.c:146
ff_dnn_flush
int ff_dnn_flush(DnnContext *ctx)
Definition: dnn_filter_common.c:166
ClassifyPostProc
int(* ClassifyPostProc)(AVFrame *frame, DNNData *output, uint32_t bbox_index, AVFilterContext *filter_ctx)
Definition: dnn_interface.h:91
DNNBackendType
DNNBackendType
Definition: dnn_interface.h:35
ff_get_dnn_module
const DNNModule * ff_get_dnn_module(DNNBackendType backend_type, void *log_ctx)
Definition: dnn_interface.c:33
input
and forward the test the status of outputs and forward it to the corresponding return FFERROR_NOT_READY If the filters stores internally one or a few frame for some input
Definition: filter_design.txt:172
i
#define i(width, name, range_min, range_max)
Definition: cbs_h2645.c:255
ff_dnn_set_frame_proc
int ff_dnn_set_frame_proc(DnnContext *ctx, FramePrePostProc pre_proc, FramePrePostProc post_proc)
Definition: dnn_filter_common.c:102
av_calloc
void * av_calloc(size_t nmemb, size_t size)
Definition: mem.c:262
ff_dnn_get_output
int ff_dnn_get_output(DnnContext *ctx, int input_width, int input_height, int *output_width, int *output_height)
Definition: dnn_filter_common.c:126
av_get_token
char * av_get_token(const char **buf, const char *term)
Unescape the given string until a non escaped terminating char, and return the token corresponding to...
Definition: avstring.c:143
FramePrePostProc
int(* FramePrePostProc)(AVFrame *frame, DNNData *model, AVFilterContext *filter_ctx)
Definition: dnn_interface.h:89
AVFilterContext
An instance of a filter.
Definition: avfilter.h:407
DNN_TH
@ DNN_TH
Definition: dnn_interface.h:35
av_free
#define av_free(p)
Definition: tableprint_vlc.h:33
ff_dnn_init
int ff_dnn_init(DnnContext *ctx, DNNFunctionType func_type, AVFilterContext *filter_ctx)
Definition: dnn_filter_common.c:54
av_freep
#define av_freep(p)
Definition: tableprint_vlc.h:34
DNNExecBaseParams
Definition: dnn_interface.h:76
av_log
#define av_log(a,...)
Definition: tableprint_vlc.h:27
ff_dnn_uninit
void ff_dnn_uninit(DnnContext *ctx)
Definition: dnn_filter_common.c:171
ff_dnn_execute_model
int ff_dnn_execute_model(DnnContext *ctx, AVFrame *in_frame, AVFrame *out_frame)
Definition: dnn_filter_common.c:134
avstring.h
DNNAsyncStatusType
DNNAsyncStatusType
Definition: dnn_interface.h:45
ff_dnn_set_classify_post_proc
int ff_dnn_set_classify_post_proc(DnnContext *ctx, ClassifyPostProc post_proc)
Definition: dnn_filter_common.c:115