동영상이 이러한 사양을 충족하지 않으면 Vertex AI Vision에서 입력을 제대로 처리하지 못할 수 있습니다.
처리 중에 오디오가 누락됩니다.
스트림이 창고에 연결된 앱의 일부인 경우 동영상 매개변수(예: FPS 또는 해상도)는 전체 스트림에서 동일해야 합니다. RTSP 동영상 데이터 매개변수 또는 로컬 동영상 데이터 매개변수의 변형은 지원되지 않습니다.
모델 시작 지연으로 인해 처리된 동영상에서 처음 몇 초의 콘텐츠가 누락될 수 있습니다. 이 시간은 최대 15초까지 걸릴 수 있습니다.
연결 시간이 길어질 경우 처리가 종료될 수 있습니다(평균 5시간). vaictl 도구에는 다시 연결 기능이 없습니다. 사용자는 수동으로 다시 연결해야 합니다.
vaictl는 동영상 데이터를 처리하는 데 유용한 도구이지만 이 도구는 네트워크 오류 복구를 자동으로 처리하지 않습니다. 이러한 오류는 데이터 소스 측 또는 Cloud 처리 측에서 발생할 수 있습니다. 예기치 않은 vaictl 작업 종료를 처리하는 다시 시작 스크립트를 만드는 것은 사용자의 책임입니다.
LOCATION_ID: 위치 ID입니다. 예를 들면 us-central1입니다.
추가 정보지원되는 리전
LOCAL_FILE.EXT: 로컬 동영상 파일의 파일 이름입니다. 예를 들면 my-video.mp4입니다.
STREAM_ID: 클러스터에서 만든 스트림 ID입니다(예: input-1).
RTSP_ADDRESS: 실시간 스트리밍 프로토콜(RTSP) 피드의 주소입니다. 예를 들면 rtsp://my-ip-camera입니다.
로컬 동영상 데이터:
# This command streams a video file to a stream. Streaming ends when the video ends.
vaictl -p PROJECT_ID \
-l LOCATION_ID \
-c application-cluster-0 \
--service-endpoint visionai.googleapis.com \
send video-file to streams STREAM_ID --file-path LOCAL_FILE.EXT
로컬 동영상 데이터 (반복):
# This command streams a video file to a stream. Video is looped into the stream until you stop the command.
vaictl -p PROJECT_ID \
-l LOCATION_ID \
-c application-cluster-0 \
--service-endpoint visionai.googleapis.com \
send video-file to streams STREAM_ID --file-path LOCAL_FILE.EXT--loop
실시간 스트리밍 프로토콜 (RTSP) 피드 데이터:
# This command will send an RTSP feed into the stream.# This command has to run in the network that has direct access to the RTSP feed.
vaictl-pPROJECT_ID\-lLOCATION_ID\-capplication-cluster-0\--service-endpointvisionai.googleapis.com\
sendrtsptostreamsSTREAM_ID--rtsp-uriRTSP_ADDRESS
처리된 동영상 재생
스트림을 만들고 동영상 콘텐츠를 스트림으로 전송한 후 동영상 콘텐츠를 재생할 수 있습니다.
스트림 보기 탭을 1시간 넘게 열어두면 페이지를 새로고침해야 동영상 콘텐츠를 계속 로드할 수 있습니다.
[[["이해하기 쉬움","easyToUnderstand","thumb-up"],["문제가 해결됨","solvedMyProblem","thumb-up"],["기타","otherUp","thumb-up"]],[["이해하기 어려움","hardToUnderstand","thumb-down"],["잘못된 정보 또는 샘플 코드","incorrectInformationOrSampleCode","thumb-down"],["필요한 정보/샘플이 없음","missingTheInformationSamplesINeed","thumb-down"],["번역 문제","translationIssue","thumb-down"],["기타","otherDown","thumb-down"]],["최종 업데이트: 2025-09-04(UTC)"],[],[],null,["# Create streams and ingest data\n\nThe first step in analyzing your video data with an application is\ncreating a pipeline for the continuous flow of data.\nStreams offers the ability to ingest real-time video data, which then\ncan be used as input for models or stored in a Warehouse.\n\nCreate a Stream\n---------------\n\nTo be able to stream video data you must first create a stream.\n\n\nThe first time you create a stream in a new project it can take some time to create the stream.\nThis time is around 30 minutes. This time is due to the system needing to initialize a\ncomputation cluster for subsequent computations.\n\n\u003cbr /\u003e\n\n### Console\n\nCreate a stream in the Google Cloud console.\n\n1. Open the **Streams** tab of the Vertex AI Vision dashboard.\n\n [Go to the Streams tab](https://console.cloud.google.com/ai/vision-ai/video-streams)\n2. Click add**Register**.\n\n3. Enter the stream name and select a region. You can click **Add Row** to\n register multiple streams at the same time.\n\n4. Click the **Register** button to create one or more streams.\n\nIngest videos\n-------------\n\nAfter you have created a stream you can begin to ingest data using that stream.\n\n\nSome limitations and considerations apply to ingested video:\n\n- The input video source must have the following specifications:\n - [H.264 encoding](https://en.wikipedia.org/wiki/Advanced_Video_Coding)\n - \\\u003c= 1080p resolution\n - \\~25\n\n If the video doesn't meet these specifications Vertex AI Vision may not\n process the input well.\n- Audio is dropped during ingestion.\n- If the stream is part of an app connected to a warehouse, video parameters (such as FPS or resolution) must be the same for the whole stream ; variations in RTSP video data parameters or local video data parameters are not supported.\n- Due to model startup latency, ingested videos may have the first few seconds of content missing. This amount of time can reach up to fifteen seconds.\n- Ingestion termination can happen after long connection (\\~5 hours, on average). There's no reconnection capability from the `vaictl` tool ; users must manually reconnect.\n- While `vaictl` is a useful tool for ingesting video data, the tool does not automatically handle recovery from network errors. These errors may either come from the data-source side or the Cloud-ingestion side. It's the user's responsibility to create a restart script to handle unexpected `vaictl` operation terminations.\n\n\u003cbr /\u003e\n\n### Vertex AI Vision SDK\n\nTo send a request to ingest video data using an existing stream you must\n[install the Vertex AI Vision SDK](/vision-ai/docs/cloud-environment#install_the_streams_sdk).\n\n\n| **Note** : You may need to establish user access credentials by running the following command: `gcloud auth application-default login`. For more information about this command, see the [Vertex AI Vision SDK\n| reference documentation](/sdk/gcloud/reference/auth/application-default/login).\n\n\u003cbr /\u003e\n\nMake the following variable substitutions:\n\n- \u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e: Your Google Cloud project ID.\n- \u003cvar translate=\"no\"\u003eLOCATION_ID\u003c/var\u003e: Your location ID. For example, `us-central1`. [More information](/about/locations). [Supported regions](/vision-ai/docs/warehouse-supported-regions).\n- \u003cvar translate=\"no\"\u003eLOCAL_FILE.EXT\u003c/var\u003e: The filename of a local video file. For example, `my-video.mp4`.\n- \u003cvar translate=\"no\"\u003eSTREAM_ID\u003c/var\u003e: The stream ID that you created in the cluster For example, `input-1`.\n- \u003cvar translate=\"no\"\u003eRTSP_ADDRESS\u003c/var\u003e: The address of your Real Time Streaming Protocol ([RTSP](https://en.wikipedia.org/wiki/Real_Time_Streaming_Protocol)) feed. For example, `rtsp://my-ip-camera`.\n\n**Local video data:** \n\n # This command streams a video file to a stream. Streaming ends when the video ends.\n vaictl -p \u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e \\\n -l \u003cvar translate=\"no\"\u003eLOCATION_ID\u003c/var\u003e \\\n -c application-cluster-0 \\\n --service-endpoint visionai.googleapis.com \\\n send video-file to streams \u003cvar translate=\"no\"\u003eSTREAM_ID\u003c/var\u003e --file-path \u003cvar translate=\"no\"\u003eLOCAL_FILE.EXT\u003c/var\u003e\n\n**Local video data (looped):** \n\n # This command streams a video file to a stream. Video is looped into the stream until you stop the command.\n vaictl -p \u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e \\\n -l \u003cvar translate=\"no\"\u003eLOCATION_ID\u003c/var\u003e \\\n -c application-cluster-0 \\\n --service-endpoint visionai.googleapis.com \\\n send video-file to streams \u003cvar translate=\"no\"\u003eSTREAM_ID\u003c/var\u003e --file-path \u003cvar translate=\"no\"\u003eLOCAL_FILE.EXT\u003c/var\u003e --loop\n\n**Real Time Streaming Protocol (RTSP) feed data:** \n\n # This command will send an RTSP feed into the stream.\n # This command has to run in the network that has direct access to the RTSP feed.\n vaictl -p \u003cvar translate=\"no\"\u003ePROJECT_ID\u003c/var\u003e \\\n -l \u003cvar translate=\"no\"\u003eLOCATION_ID\u003c/var\u003e \\\n -c application-cluster-0 \\\n --service-endpoint visionai.googleapis.com \\\n send rtsp to streams \u003cvar translate=\"no\"\u003eSTREAM_ID\u003c/var\u003e --rtsp-uri \u003cvar translate=\"no\"\u003eRTSP_ADDRESS\u003c/var\u003e\n\nPlay ingested videos\n--------------------\n\nAfter you have created a stream and sent video content to the stream you can\nplay the video content.\n\nIf you leave the stream view tab open for more than one hour, you must\nrefresh the page to continue to load video contents. \n\n### Console\n\nPlay ingested video in the Google Cloud console.\n\n1. Open the **Streams** tab of the Vertex AI Vision dashboard.\n\n [Go to the Streams tab](https://console.cloud.google.com/ai/vision-ai/video-streams)\n2. Click on the stream name from the streams list.\n\n3. In the video stream detail page the ingested video live view\n automatically loads."]]
