WebRTC VideoEngine综合应用示例(一)——视频通话的基本流程
本系列目前共三篇文章,后续还会更新
WebRTC VideoEngine综合应用示例(一)——视频通话的基本流程
WebRTC VideoEngine综合应用示例(二)——集成OPENH264编解码器
WebRTC VideoEngine综合应用示例(三)——集成X264编码和ffmpeg解码
关注下方公众号,回复“webrtc视频通话”,查看源码地址,是一个可以脱离webrtc那个大项目而独立运行的工程
关注公众号,掌握更多多媒体领域知识与资讯
文章帮到你了?可以扫描如下二维码进行打赏~,打赏多少您随意~
WebRTC技术的出现改变了传统即时通信的现状,它是一套开源的旨在建立浏览器端对端的通信标准的技术,支持浏览器平台,使用P2P架构。WebRTC所采用的技术都是当前VoIP先进的技术,如内部所采用的音频引擎是Google收购知名GIPS公司获得的核心技术:视频编解码则采用了VP8。
大家都说WebRTC好,是未来的趋势,但是不得不说这个开源项目对新手学习实在是太不友好,光是windows平台下的编译就能耗费整整一天的精力,还未必能成功,关于这个问题在我之前的文章中有所描述。编译成功之后打开一看,整个solution里面有215个项目,绝对让人当时就懵了,而且最重要的是,google方面似乎没给出什么有用的文档供人参考,网络上有关的资料也多是有关于web端开发的,和Native API开发有关的内容少之又少,于是我决定把自己这两天学习VideoEngine的成果分享出来,供大家参考,有什么问题也欢迎大家指出,一起学习一起进步。
首先需要说明的是,webrtc项目的all.sln下有一个vie_auto_test项目,里面包含了一些针对VideoEngine的测试程序,我这里的demo就是基于此修改得到的。
先来看一下VideoEngine的核心API,基本上就在以下几个头文件中了。
具体来说
ViEBase用于
- 创建和销毁 VideoEngine 实例
- 创建和销毁 channels
- 将 video channel 和相应的 voice channel 连接到一起并同步
- 发送和接收的开始与停止
ViECapture用于
- 分配capture devices.
- 将 capture device 与一个或多个 channels连接起来.
- 启动或停止 capture devices.
- 获得capture device 的可用性.
ViECodec用于
- 设置发送和接收的编解码器.
- 设置编解码器特性.
- Key frame signaling.
- Stream management settings.
ViEError即一些预定义的错误消息
ViEExternalCodec用于注册除VP8之外的其他编解码器
ViEImageProcess提供以下功能
- Effect filters
- 抗闪烁
- 色彩增强
ViENetwork用于
- 配置发送和接收地址.
- External transport support.
- 端口和地址过滤.
- Windows GQoS functions and ToS functions.
- Packet timeout notification.
- Dead‐or‐Alive connection observations.
ViERender用于
- 为输入视频流、capture device和文件指定渲染目标.
- 配置render streams.
ViERTP_RTCP用于
- Callbacks for RTP and RTCP events such as modified SSRC or CSRC.
- SSRC handling.
- Transmission of RTCP reports.
- Obtaining RTCP data from incoming RTCP sender reports.
- RTP and RTCP statistics (jitter, packet loss, RTT etc.).
- Forward Error Correction (FEC).
- Writing RTP and RTCP packets to binary files for off‐line analysis of the call quality.
- Inserting extra RTP packets into active audio stream.
下面将以实现一个视频通话功能为实例详细介绍VideoEngine的使用,在文末将附上相应源码的下载地址
第一步是创建一个VideoEngine实例,如下
webrtc::VideoEngine* ptrViE = NULL;ptrViE = webrtc::VideoEngine::Create();if (ptrViE == NULL){printf("ERROR in VideoEngine::Create\n");return -1;}
然后初始化VideoEngine并创建一个Channel
webrtc::ViEBase* ptrViEBase = webrtc::ViEBase::GetInterface(ptrViE);if (ptrViEBase == NULL){printf("ERROR in ViEBase::GetInterface\n");return -1;}error = ptrViEBase->Init();//这里的Init其实是针对VideoEngine的初始化if (error == -1){printf("ERROR in ViEBase::Init\n");return -1;}webrtc::ViERTP_RTCP* ptrViERtpRtcp =webrtc::ViERTP_RTCP::GetInterface(ptrViE);if (ptrViERtpRtcp == NULL){printf("ERROR in ViERTP_RTCP::GetInterface\n");return -1;}int videoChannel = -1;error = ptrViEBase->CreateChannel(videoChannel);if (error == -1){printf("ERROR in ViEBase::CreateChannel\n");return -1;}
列出可用的capture devices等待用户进行选择, 然后进行allocate和connect,最后start选中的capture device
webrtc::ViECapture* ptrViECapture =webrtc::ViECapture::GetInterface(ptrViE);if (ptrViEBase == NULL){printf("ERROR in ViECapture::GetInterface\n");return -1;}const unsigned int KMaxDeviceNameLength = 128;const unsigned int KMaxUniqueIdLength = 256;char deviceName[KMaxDeviceNameLength];memset(deviceName, 0, KMaxDeviceNameLength);char uniqueId[KMaxUniqueIdLength];memset(uniqueId, 0, KMaxUniqueIdLength);printf("Available capture devices:\n");int captureIdx = 0;for (captureIdx = 0;captureIdx < ptrViECapture->NumberOfCaptureDevices();captureIdx++){memset(deviceName, 0, KMaxDeviceNameLength);memset(uniqueId, 0, KMaxUniqueIdLength);error = ptrViECapture->GetCaptureDevice(captureIdx, deviceName,KMaxDeviceNameLength, uniqueId,KMaxUniqueIdLength);if (error == -1){printf("ERROR in ViECapture::GetCaptureDevice\n");return -1;}printf("\t %d. %s\n", captureIdx + 1, deviceName);}printf("\nChoose capture device: ");if (scanf("%d", &captureIdx) != 1){printf("Error in scanf()\n");return -1;}getchar();captureIdx = captureIdx - 1; // Compensate for idx start at 1.error = ptrViECapture->GetCaptureDevice(captureIdx, deviceName,KMaxDeviceNameLength, uniqueId,KMaxUniqueIdLength);if (error == -1){printf("ERROR in ViECapture::GetCaptureDevice\n");return -1;}int captureId = 0;error = ptrViECapture->AllocateCaptureDevice(uniqueId, KMaxUniqueIdLength,captureId);if (error == -1){printf("ERROR in ViECapture::AllocateCaptureDevice\n");return -1;}error = ptrViECapture->ConnectCaptureDevice(captureId, videoChannel);if (error == -1){printf("ERROR in ViECapture::ConnectCaptureDevice\n");return -1;}error = ptrViECapture->StartCapture(captureId);if (error == -1){printf("ERROR in ViECapture::StartCapture\n");return -1;}
设置RTP/RTCP所采用的模式
error = ptrViERtpRtcp->SetRTCPStatus(videoChannel,webrtc::kRtcpCompound_RFC4585);if (error == -1){printf("ERROR in ViERTP_RTCP::SetRTCPStatus\n");return -1;}
设置接收端解码器出问题的时候,比如关键帧丢失或损坏,如何重新请求关键帧的方式
error = ptrViERtpRtcp->SetKeyFrameRequestMethod(videoChannel, webrtc::kViEKeyFrameRequestPliRtcp);if (error == -1){printf("ERROR in ViERTP_RTCP::SetKeyFrameRequestMethod\n");return -1;}
设置是否为当前channel使用REMB(Receiver Estimated Max Bitrate)包,发送端可以用它表明正在编码当前channel
接收端用它来记录当前channel的估计码率
error = ptrViERtpRtcp->SetRembStatus(videoChannel, true, true);if (error == -1){printf("ERROR in ViERTP_RTCP::SetTMMBRStatus\n");return -1;}
设置rendering用于显示
webrtc::ViERender* ptrViERender = webrtc::ViERender::GetInterface(ptrViE);if (ptrViERender == NULL) {printf("ERROR in ViERender::GetInterface\n");return -1;}
显示本地摄像头数据,这里的window1和下面的window2都是显示窗口,更详细的内容后面再说
error= ptrViERender->AddRenderer(captureId, window1, 0, 0.0, 0.0, 1.0, 1.0);if (error == -1){printf("ERROR in ViERender::AddRenderer\n");return -1;}error = ptrViERender->StartRender(captureId);if (error == -1){printf("ERROR in ViERender::StartRender\n");return -1;}
显示接收端收到的解码数据
error = ptrViERender->AddRenderer(videoChannel, window2, 1, 0.0, 0.0, 1.0,1.0);if (error == -1){printf("ERROR in ViERender::AddRenderer\n");return -1;}error = ptrViERender->StartRender(videoChannel);if (error == -1){printf("ERROR in ViERender::StartRender\n");return -1;}
设置编解码器
webrtc::ViECodec* ptrViECodec = webrtc::ViECodec::GetInterface(ptrViE);if (ptrViECodec == NULL){printf("ERROR in ViECodec::GetInterface\n");return -1;}VideoCodec videoCodec;int numOfVeCodecs = ptrViECodec->NumberOfCodecs();for (int i = 0; i<numOfVeCodecs; ++i){if (ptrViECodec->GetCodec(i, videoCodec) != -1){if (videoCodec.codecType == kVideoCodecVP8)break;}}videoCodec.targetBitrate = 256;videoCodec.minBitrate = 200;videoCodec.maxBitrate = 300;videoCodec.maxFramerate = 25;error = ptrViECodec->SetSendCodec(videoChannel, videoCodec);assert(error != -1);error = ptrViECodec->SetReceiveCodec(videoChannel, videoCodec);assert(error != -1);
设置接收和发送地址,然后开始发送和接收
webrtc::ViENetwork* ptrViENetwork =webrtc::ViENetwork::GetInterface(ptrViE);if (ptrViENetwork == NULL){printf("ERROR in ViENetwork::GetInterface\n");return -1;}//VideoChannelTransport是由我们自己定义的类,后面将会详细介绍VideoChannelTransport* video_channel_transport = NULL;video_channel_transport = new VideoChannelTransport(ptrViENetwork, videoChannel);const char* ipAddress = "127.0.0.1";const unsigned short rtpPort = 6000;std::cout << std::endl;std::cout << "Using rtp port: " << rtpPort << std::endl;std::cout << std::endl;error = video_channel_transport->SetLocalReceiver(rtpPort);if (error == -1){printf("ERROR in SetLocalReceiver\n");return -1;}error = video_channel_transport->SetSendDestination(ipAddress, rtpPort);if (error == -1){printf("ERROR in SetSendDestination\n");return -1;}error = ptrViEBase->StartReceive(videoChannel);if (error == -1){printf("ERROR in ViENetwork::StartReceive\n");return -1;}error = ptrViEBase->StartSend(videoChannel);if (error == -1){printf("ERROR in ViENetwork::StartSend\n");return -1;}
设置按下回车键即停止通话
printf("\n call started\n\n");printf("Press enter to stop...");while ((getchar()) != '\n');
停止通话后的各种stop
error = ptrViEBase->StopReceive(videoChannel);if (error == -1){printf("ERROR in ViEBase::StopReceive\n");return -1;}error = ptrViEBase->StopSend(videoChannel);if (error == -1){printf("ERROR in ViEBase::StopSend\n");return -1;}error = ptrViERender->StopRender(captureId);if (error == -1){printf("ERROR in ViERender::StopRender\n");return -1;}error = ptrViERender->RemoveRenderer(captureId);if (error == -1){printf("ERROR in ViERender::RemoveRenderer\n");return -1;}error = ptrViERender->StopRender(videoChannel);if (error == -1){printf("ERROR in ViERender::StopRender\n");return -1;}error = ptrViERender->RemoveRenderer(videoChannel);if (error == -1){printf("ERROR in ViERender::RemoveRenderer\n");return -1;}error = ptrViECapture->StopCapture(captureId);if (error == -1){printf("ERROR in ViECapture::StopCapture\n");return -1;}error = ptrViECapture->DisconnectCaptureDevice(videoChannel);if (error == -1){printf("ERROR in ViECapture::DisconnectCaptureDevice\n");return -1;}error = ptrViECapture->ReleaseCaptureDevice(captureId);if (error == -1){printf("ERROR in ViECapture::ReleaseCaptureDevice\n");return -1;}error = ptrViEBase->DeleteChannel(videoChannel);if (error == -1){printf("ERROR in ViEBase::DeleteChannel\n");return -1;}delete video_channel_transport;int remainingInterfaces = 0;remainingInterfaces = ptrViECodec->Release();remainingInterfaces += ptrViECapture->Release();remainingInterfaces += ptrViERtpRtcp->Release();remainingInterfaces += ptrViERender->Release();remainingInterfaces += ptrViENetwork->Release();remainingInterfaces += ptrViEBase->Release();if (remainingInterfaces > 0){printf("ERROR: Could not release all interfaces\n");return -1;}bool deleted = webrtc::VideoEngine::Delete(ptrViE);if (deleted == false){printf("ERROR in VideoEngine::Delete\n");return -1;}return 0;
以上就是VideoEngine的基本使用流程,下面说一下显示窗口如何创建
这里使用了webrtc已经为我们定义好的类ViEWindowCreator,它有一个成员函数CreateTwoWindows可以直接创建两个窗口,只需实现定义好窗口名称、窗口大小以及坐标即可,如下
ViEWindowCreator windowCreator;ViEAutoTestWindowManagerInterface* windowManager =windowCreator.CreateTwoWindows();VideoEngineSample(windowManager->GetWindow1(),windowManager->GetWindow2());
这里的VideoEngineSample就是我们在前面所写的包含全部流程的示例程序,它以两个窗口的指针作为参数
至于前面提到的VideoChannelTransport定义如下
class VideoChannelTransport : public webrtc::test::UdpTransportData {
public:VideoChannelTransport(ViENetwork* vie_network, int channel);virtual ~VideoChannelTransport();// Start implementation of UdpTransportData.virtual void IncomingRTPPacket(const int8_t* incoming_rtp_packet,const int32_t packet_length,const char* /*from_ip*/,const uint16_t /*from_port*/) OVERRIDE;virtual void IncomingRTCPPacket(const int8_t* incoming_rtcp_packet,const int32_t packet_length,const char* /*from_ip*/,const uint16_t /*from_port*/) OVERRIDE;// End implementation of UdpTransportData.// Specifies the ports to receive RTP packets on.int SetLocalReceiver(uint16_t rtp_port);// Specifies the destination port and IP address for a specified channel.int SetSendDestination(const char* ip_address, uint16_t rtp_port);private:int channel_;ViENetwork* vie_network_;webrtc::test::UdpTransport* socket_transport_;
};VideoChannelTransport::VideoChannelTransport(ViENetwork* vie_network,int channel): channel_(channel),vie_network_(vie_network) {uint8_t socket_threads = 1;socket_transport_ = webrtc::test::UdpTransport::Create(channel, socket_threads);int registered = vie_network_->RegisterSendTransport(channel,*socket_transport_);
}VideoChannelTransport::~VideoChannelTransport() {vie_network_->DeregisterSendTransport(channel_);webrtc::test::UdpTransport::Destroy(socket_transport_);
}void VideoChannelTransport::IncomingRTPPacket(const int8_t* incoming_rtp_packet,const int32_t packet_length,const char* /*from_ip*/,const uint16_t /*from_port*/) {vie_network_->ReceivedRTPPacket(channel_, incoming_rtp_packet, packet_length, PacketTime());
}void VideoChannelTransport::IncomingRTCPPacket(const int8_t* incoming_rtcp_packet,const int32_t packet_length,const char* /*from_ip*/,const uint16_t /*from_port*/) {vie_network_->ReceivedRTCPPacket(channel_, incoming_rtcp_packet,packet_length);
}int VideoChannelTransport::SetLocalReceiver(uint16_t rtp_port) {int return_value = socket_transport_->InitializeReceiveSockets(this,rtp_port);if (return_value == 0) {return socket_transport_->StartReceiving(500);}return return_value;
}int VideoChannelTransport::SetSendDestination(const char* ip_address,uint16_t rtp_port) {return socket_transport_->InitializeSendSockets(ip_address, rtp_port);
}
继承自UdpTransportData类,主要重写了IncomingRTPPacket和IncomingRTCPPacket两个成员函数,分别调用了vie_network的ReceivedRTPPacket和ReceivedRTCPPacket方法,当需要将接收到的RTP和RTCP包传给VideoEngine时就应该使用这两个函数。
该示例程序最后效果如下,我这里是几个虚拟摄像头,然后会有两个窗口,一个是摄像头画面,一个是解码的画面。
WebRTC VideoEngine综合应用示例(一)——视频通话的基本流程相关推荐
- WebRTC VoiceEngine综合应用示例(一)——基本结构分析
转自 zhanghui_cuc :http://blog.csdn.net/nonmarking/article/details/50574733 把自己这两天学习VoiceEngine的成果分享出来 ...
- WebRTC VideoEngine超详细教程(三)——集成X264编码和ffmpeg解码
转自:http://blog.csdn.net/nonmarking/article/details/47958395 本系列目前共三篇文章,后续还会更新 WebRTC VideoEngine超详细教 ...
- webrtc 入门第五章 一对一视频通话实现
webrtc 入门第五章 一对一视频通话实现 一.介绍 在前面的章节我们学习了如何操作本地的设备摄像头,麦克风等,学会了如何进行本地的流媒体操作如录制,下载,同步等.在第三第四章节学习了webrt ...
- Cisco Easy ***综合配置示例
以下内容摘自正在全面热销的最新网络设备图书"豪华四件套"之一<Cisco路由器配置与管理完全手册>(第二版)(其余三本分别是:<Cisco交换机配置与管理完全手册 ...
- webrtc入门:14.pion webrtc中Data Channels示例
Data Channels 在pion webrtc 中有非常多的示例,Data Channels 就是其中的一个,当我们第一次打开pion webrtc的示例时,可能会有点不知所措,不知道他要让我们 ...
- 案例1:使用awk提取文本案例2:awk处理条件案例3:awk综合脚本应用案例4:awk流程控制案例5:awk扩展应用
案例1:使用awk提取文本 案例2:awk处理条件 案例3:awk综合脚本应用 案例4:awk流程控制 案例5:awk扩展应用 1 案例1:使用awk提取文本 1.1 问题 本案例要求使用awk工具完 ...
- 结合WebSocket编写WebGL综合场景示例
在WebGL场景中导入多个Babylon骨骼模型,在局域网用WebSocket实现多用户交互控制. 首先是场景截图: 上图在场景中导入一个Babylon骨骼模型,使用asdw.空格.鼠标控制加速度移动 ...
- 一文带你了解webrtc基本原理(动手实现1v1视频通话)
webrtc (Web Real-Time Communications) 是一个实时通讯技术,也是实时音视频技术的标准和框架. 大白话讲,webrtc是一个集大成的实时音视频技术集,包含了各种客户端 ...
- ffmpeg综合应用示例(一)——摄像头直播
本文的示例将实现:读取PC摄像头视频数据并以RTMP协议发送为直播流.示例包含了 1.ffmpeg的libavdevice的使用 2.视频解码.编码.推流的基本流程 具有较强的综合性. 要使用liba ...
- ffmpeg综合应用示例(二)——为直播流添加特效
在上一篇文章中,讲解了如何利用ffmpeg实现摄像头直播,本文将在此基础上,实现一个可以选择各种视频滤镜的摄像头直播示例.本文包含以下内容 1.AVFilter的基本介绍 2.如何利用ffmpeg命令 ...
最新文章
- E:By Elevator or Stairs? CF595 DP最短路
- 怎么获得combobox的valueField值
- IntelliJ 创建main函数快捷
- Linux下Sniffer程序的实现
- Zedboard学习(三):PL下流水灯实验
- sscom 中文显示 乱码_解决SSM框架使用过程中的中文乱码问题
- 七夕过后,你分析过自己单身原因的原因吗?
- dev 报表设计器 怎么设置每页10行_可嵌入您系统的.NET 报表控件ActiveReports:带状列表组件...
- ARC106E-Medals【hall定理,高维前缀和】
- 猪八戒网的DevOps进化论
- Android,EditText,InuputType
- 「开源资讯」Sentinel Go 0.4.0 发布,支持热点流量防护能力
- 如何提高自己的象棋水平及象棋开局的五种忌讳
- Ansys-自适应网格划分-受压薄板学习收获
- 埃森哲互动并购了56家广告公司
- 移动架构11_建造者模式
- 自媒体必做的一个平台
- 解析ARM中OS_CPU_A.S(中断级方式)
- C++指针详解2_typedef函数声明类型、sizeof特性简介与数组指针间关系说明
- Python-使用正则表达式爬取斗破苍穹小说文字内容(使用Requests库实现)