完全兼容 FFmpeg
完全兼容 FFmpeg
BMF 与 FFmpeg 的能力完全兼容。FFmpeg 以其丰富的 AV 解封装、解码、filter、编码、封装等能力而闻名业界。 BMF 为 Python、C++、Go 用户提供了内置模块和灵活的 API，以便将 FFmpeg 能力应用到自己的解决方案中。
环境
需要使用 FFmpeg 库，支持的版本和安装详情可在开始使用->安装部署中找到。
能力
- demux + decode
- filter
- encode + mux
详细参数可以在 API ->内置解码模块，API ->内置 filter 模块和 API ->内置编码模块中找到。
如何使用
下面用一些经典的示例来展示在多媒体处理场景中使用 BMF 的方法。
仅解码
下方示例仅解码了一个多媒体文件：
import bmf
input_video_path = "../files/test.mp4"
graph = bmf.graph()
stream = graph.decode({
"input_path": input_video_path
})
(
bmf.encode(
stream['video'],
stream['audio'],
{
"null_output": 1
}
)
.run()
)
转码
为了实现模块并行，编码模块将应用于 scheduler 1，而解码和其他模块默认应用于 scheduler 0。
import bmf
input_video_path = "test.mp4"
output_path = "./output.mp4"
graph = bmf.graph({'dump_graph': 1})
stream = graph.decode({
"input_path": input_video_path,
"dec_params": {
"threads": "8"
}
})
# using scale filter
scaled = stream['video'].scale(320, 240)
(
bmf.encode(
scaled,
stream['audio'],
{
"output_path": output_path,
"video_params": {
"codec": "h264",
"width": 320,
"height": 240,
"crf": 23,
"preset": "veryfast",
},
"audio_params": {
"codec": "aac",
"bit_rate": 128000,
"sample_rate": 44100,
"channels": 2
}
}
).run()
)
BMF 也支持 FFmpeg GPU 编解码器。例如，您可以设置 "hwaccel"： "cuda" 和 "codec"： "hevc_nvenc" 来使用 GPU 进行解码和编码。详情请参阅 GPU 转码章节。
图像编码
import bmf
input_video_path = "test.png"
output_path = "./image.jpg"
(
bmf.graph()
.decode({'input_path': input_video_path})['video']
.scale(320, 240)
.encode(None,
{
"output_path": output_path,
"format": "mjpeg",
"video_params": {
"codec": "jpg",
"width": 320,
"height": 240
}
}
).run()
)
复制 Stream
import bmf
input_path = "test.mp4"
output_path = "./stream_copy.mp4"
stream = bmf.graph().decode(
{
'input_path': input_path,
'video_codec': "copy"
}
)
video_stream = stream['video']
video_stream.encode(stream['audio'], {
"output_path": "stream_copy.mp4",
}).run()
通过参数使用 FFmpeg Filter
import bmf
input_video_path1 = "test1.mp4"
input_video_path2 = "test2.mp4"
graph = bmf.graph({'dump_graph':1})
stream1 = graph.decode({
"input_path": input_video_path1
})
stream2 = graph.decode({
"input_path": input_video_path2
})
#using "vstack" ffmpeg filter in a common way
bmf.ff_filter([stream1, stream2], 'vstack', input=2).encode(None, {"output_path": "output.mp4"})
graph.run()
BMF 还支持 ffmpeg CUDA fliter，调用 ffmpeg CUDA fliter 和调用 CPU fliter 非常相似，只需要注意数据的位置。详情请参阅 gpu filter 章节。
直接使用模块能力（Sync Mode）
用户可以将模块的能力集成到自己的项目中。例如，通过调用 encode() 来解码 yuv 帧或编码 yuv 帧.
import bmf
input_video_path = "test.png"
output_path = "output.jpg"
# create decoder
decoder = bmf_sync.sync_module("c_ffmpeg_decoder", {"input_path": input_video_path}, [], [0])
'''
# for non-builtin modules, use module_info instead of module_name to specify type/path/entry
module_info = {
"name": "my_module",
"type": "",
"path": "",
"entry": ""
}
module = bmf_sync.sync_module(module_info, {"input_path": input_video_path}, [], [0])
'''
# create scale
scale = bmf_sync.sync_module("c_ffmpeg_filter", {
"name": "scale", "para": "320:240"
}, [0], [0])
# create encoder
encoder = bmf_sync.sync_module("c_ffmpeg_encoder", {
"output_path": output_path,
"format": "mjpeg",
"video_params": {
"codec": "jpg"
}
}, [0], [])
# call init if necessary, otherwise we skip this step
decoder.init()
scale.init()
encoder.init()
# decode
frames, _ = bmf_sync.process(decoder, None)
# scale frames, _ = bmf_sync.process(scale, {0:frames[0]})
# encode
bmf_sync.process(encoder, {0:frames[0]})
# send eof to encoder
bmf_sync.send_eof(encoder)
# call close if necessary, otherwise we skip this step
decoder.close() scale.close() encoder.close()
其它参考
在 test_transcode.py、test_sync_mode.py 等文件中也有很多示例。如有需要，请参考这些示例代码。
工具
BMF 提供了一些有用的工具，帮助开发人员进行调试、比较和快速验证等。
运行 Graph
app 使用 “{‘dump_graph’: 1}” 参数运行后，例如在转码中，json 描述将输出到文件 original_graph.json 中，如下所示：
{
"input_streams": [],
"output_streams": [],
"nodes": [
{
"module_info": {
"name": "c_ffmpeg_decoder",
"type": "",
"path": "",
"entry": ""
},
"meta_info": {
"premodule_id": -1,
"callback_binding": []
},
"option": {
"input_path": "test.mp4",
"dec_params": {
"threads": "8"
}
},
"input_streams": [],
"output_streams": [
{
"identifier": "video:c_ffmpeg_decoder_0_1",
"stream_alias": ""
},
{
"identifier": "audio:c_ffmpeg_decoder_0_2",
"stream_alias": ""
}
],
"input_manager": "immediate",
"scheduler": 0,
"alias": "",
"id": 0
},
{
"module_info": {
"name": "c_ffmpeg_filter",
"type": "",
"path": "",
"entry": ""
},
"meta_info": {
"premodule_id": -1,
"callback_binding": []
},
"option": {
"name": "scale",
"para": "320:240"
},
"input_streams": [
{
"identifier": "c_ffmpeg_decoder_0_1",
"stream_alias": ""
}
],
"output_streams": [
{
"identifier": "c_ffmpeg_filter_1_0",
"stream_alias": ""
}
],
"input_manager": "immediate",
"scheduler": 0,
"alias": "",
"id": 1
},
{
"module_info": {
"name": "c_ffmpeg_encoder",
"type": "",
"path": "",
"entry": ""
},
"meta_info": {
"premodule_id": -1,
"callback_binding": []
},
"option": {
"name": "scale",
"para": "320:240"
},
"input_streams": [
{
"identifier": "c_ffmpeg_decoder_0_1",
"stream_alias": ""
}
],
"output_streams": [
{
"identifier": "c_ffmpeg_filter_1_0",
"stream_alias": ""
}
],
"input_manager": "immediate",
"scheduler": 0,
"alias": "",
"id": 1
},
{
"module_info": {
"name": "c_ffmpeg_encoder",
"type": "",
"path": "",
"entry": ""
},
"meta_info": {
"premodule_id": -1,
"callback_binding": []
},
"option": {
"output_path": "./output.mp4",
"video_params": {
"codec": "h264",
"width": 320,
"height": 240,
"crf": 23,
"preset": "veryfast"
},
"audio_params": {
"codec": "aac",
"bit_rate": 128000,
"sample_rate": 44100,
"channels": 2
}
},
"input_streams": [
{
"identifier": "c_ffmpeg_filter_1_0",
"stream_alias": ""
},
{
"identifier": "c_ffmpeg_decoder_0_2",
"stream_alias": ""
}
],
"output_streams": [],
"input_manager": "immediate",
"scheduler": 1,
"alias": "",
"id": 2
}
],
"option": {
"dump_graph": 1
},
"mode": "Normal"
}
如果您知道自己正在做什么，可以查看一些细节并进行修改，还可以直接运行 json graph：
$ run_bmf_graph original_graph.json
转换为 FFmpeg 命令行
import bmf
def run():
my_graph = bmf.graph()
my_graph.runFFmpegByConfig("original_graph.json")
run()
FFmpeg 命令行将根据 json 描述生成并运行：
ffmpeg -threads 8 -i test.mp4 -filter_complex "[0:v] scale=320:240[c_ffmpeg_filter_1_0]
" -map '[c_ffmpeg_filter_1_0]' -vcodec libx264 -pix_fmt yuv420p -crf 23 -preset veryfast -s 320x240 -map
0:a -acodec aac -b:a 128000 -ar 44100 -ac 2 -f mp4 ./output.mp4 -y
最后修改
February 19, 2024
: fix baidu statistics (daa0758)