Cloud Run, Video Intelligence API, और Vertex AI का इस्तेमाल करके, वीडियो के अलग-अलग सीन की इमेज के बारे में जानकारी देने वाली सेवा बनाएं

1. परिचय

खास जानकारी

इस कोडलैब में, Node.js में लिखी गई Cloud Run सेवा बनाई जाएगी. यह सेवा, वीडियो के हर सीन की जानकारी देगी. सबसे पहले, आपकी सेवा Video Intelligence API का इस्तेमाल करके, सीन बदलने के हर समय के लिए टाइमस्टैंप का पता लगाएगी. इसके बाद, आपकी सेवा तीसरे पक्ष के बाइनरी ffmpeg का इस्तेमाल करेगी. इससे सीन में बदलाव होने के हर टाइमस्टैंप के लिए स्क्रीनशॉट कैप्चर किया जा सकेगा. आखिर में, Vertex AI की विज़ुअल कैप्शनिंग सुविधा का इस्तेमाल करके, स्क्रीनशॉट की विज़ुअल जानकारी दी जाती है.

इस कोडलैब में यह भी दिखाया गया है कि Cloud Run सेवा में ffmpeg का इस्तेमाल करके, किसी वीडियो से किसी टाइमस्टैंप पर इमेज कैसे कैप्चर की जाती हैं. ffmpeg को अलग से इंस्टॉल करना होता है. इसलिए, इस कोडलैब में आपको Dockerfile बनाने का तरीका बताया गया है. इससे ffmpeg को Cloud Run सेवा के हिस्से के तौर पर इंस्टॉल किया जा सकता है.

Cloud Run सेवा के काम करने के तरीके को यहां दिखाया गया है:

Cloud Run Video Description Service का डायग्राम

आपको क्या सीखने को मिलेगा

  • तीसरे पक्ष की बाइनरी इंस्टॉल करने के लिए, Dockerfile का इस्तेमाल करके कंटेनर इमेज बनाने का तरीका
  • Cloud Run सेवा के लिए सेवा खाता बनाकर, कम से कम विशेषाधिकार के सिद्धांत का पालन कैसे करें, ताकि वह अन्य Google Cloud सेवाओं को कॉल कर सके
  • Cloud Run सेवा से Video Intelligence क्लाइंट लाइब्रेरी का इस्तेमाल करने का तरीका
  • Vertex AI से हर सीन की विज़ुअल जानकारी पाने के लिए, Google API को कॉल करने का तरीका

2. सेटअप और ज़रूरी शर्तें

ज़रूरी शर्तें

Cloud Shell चालू करें

  1. Cloud Console में, Cloud Shell चालू करें d1264ca30785e435.png पर क्लिक करें.

cb81e7c8e34bc8d.png

अगर आपने Cloud Shell को पहली बार शुरू किया है, तो आपको एक इंटरमीडिएट स्क्रीन दिखेगी. इसमें Cloud Shell के बारे में जानकारी दी गई होगी. अगर आपको इंटरमीडिएट स्क्रीन दिखती है, तो जारी रखें पर क्लिक करें.

d95252b003979716.png

Cloud Shell से कनेक्ट होने में कुछ ही सेकंड लगेंगे.

7833d5e1c5d18f54.png

इस वर्चुअल मशीन में, डेवलपमेंट के लिए ज़रूरी सभी टूल पहले से मौजूद हैं. यह 5 जीबी की होम डायरेक्ट्री उपलब्ध कराता है, जो हमेशा बनी रहती है. साथ ही, यह Google Cloud में काम करता है. इससे नेटवर्क की परफ़ॉर्मेंस और पुष्टि करने की प्रोसेस बेहतर होती है. इस कोडलैब में ज़्यादातर काम ब्राउज़र से किया जा सकता है.

Cloud Shell से कनेक्ट होने के बाद, आपको दिखेगा कि आपकी पुष्टि हो गई है और प्रोजेक्ट को आपके प्रोजेक्ट आईडी पर सेट कर दिया गया है.

  1. पुष्टि करें कि आपने Cloud Shell में पुष्टि कर ली है. इसके लिए, यह कमांड चलाएं:
gcloud auth list

कमांड आउटपुट

 Credentialed Accounts
ACTIVE  ACCOUNT
*       <my_account>@<my_domain.com>

To set the active account, run:
    $ gcloud config set account `ACCOUNT`
  1. यह पुष्टि करने के लिए कि gcloud कमांड को आपके प्रोजेक्ट के बारे में पता है, Cloud Shell में यह कमांड चलाएं:
gcloud config list project

कमांड आउटपुट

[core]
project = <PROJECT_ID>

अगर ऐसा नहीं है, तो इस कमांड का इस्तेमाल करके इसे सेट किया जा सकता है:

gcloud config set project <PROJECT_ID>

कमांड आउटपुट

Updated property [core/project].

3. एपीआई चालू करना और एनवायरमेंट वैरिएबल सेट करना

इस कोडलैब का इस्तेमाल करने से पहले, आपको कई एपीआई चालू करने होंगे. इस कोडलैब के लिए, इन एपीआई का इस्तेमाल करना ज़रूरी है. इन एपीआई को चालू करने के लिए, यह कमांड चलाएं:

gcloud services enable run.googleapis.com \
    storage.googleapis.com \
    cloudbuild.googleapis.com \
    videointelligence.googleapis.com \
    aiplatform.googleapis.com

इसके बाद, ऐसे एनवायरमेंट वैरिएबल सेट किए जा सकते हैं जिनका इस्तेमाल इस कोडलैब में किया जाएगा.

REGION=<YOUR-REGION>
PROJECT_ID=<YOUR-PROJECT-ID>

PROJECT_NUMBER=$(gcloud projects describe $PROJECT_ID --format='value(projectNumber)')
SERVICE_NAME=video-describer
export BUCKET_ID=$PROJECT_ID-video-describer

4. Cloud Storage बकेट बनाना

Cloud Storage बकेट बनाएं. इसमें Cloud Run सेवा के ज़रिए प्रोसेस करने के लिए वीडियो अपलोड किए जा सकते हैं. इसके लिए, यह कमांड इस्तेमाल करें:

gsutil mb -l us-central1 gs://$BUCKET_ID/

[ज़रूरी नहीं] इस सैंपल वीडियो का इस्तेमाल किया जा सकता है. इसके लिए, इसे अपने डिवाइस पर डाउनलोड करें.

gsutil cp gs://cloud-samples-data/video/visionapi.mp4 testvideo.mp4

अब अपनी वीडियो फ़ाइल को स्टोरेज बकेट में अपलोड करें.

FILENAME=<YOUR-VIDEO-FILENAME>
gsutil cp $FILENAME gs://$BUCKET_ID

5. Node.js ऐप्लिकेशन बनाना

सबसे पहले, सोर्स कोड के लिए एक डायरेक्ट्री बनाएं और उस डायरेक्ट्री में cd करें.

mkdir video-describer && cd $_

इसके बाद, package.json फ़ाइल बनाएं और उसमें यह कॉन्टेंट शामिल करें:

{
  "name": "video-describer",
  "version": "1.0.0",
  "private": true,
  "description": "describes the image in every scene for a given video",
  "main": "index.js",
  "author": "Google LLC",
  "license": "Apache-2.0",
  "scripts": {
    "start": "node index.js"
  },
  "dependencies": {
    "@google-cloud/storage": "^7.7.0",
    "@google-cloud/video-intelligence": "^5.0.1",
    "axios": "^1.6.2",
    "express": "^4.18.2",
    "fluent-ffmpeg": "^2.1.2",
    "google-auth-library": "^9.4.1"
  }
}

इस ऐप्लिकेशन में कई सोर्स फ़ाइलें शामिल हैं, ताकि इसे आसानी से पढ़ा जा सके. सबसे पहले, नीचे दिए गए कॉन्टेंट के साथ index.js सोर्स फ़ाइल बनाएं. इस फ़ाइल में, सेवा के लिए एंट्री पॉइंट होता है. साथ ही, इसमें ऐप्लिकेशन के लिए मुख्य लॉजिक होता है.

const { captureImages } = require('./imageCapture.js');
const { detectSceneChanges } = require('./sceneDetector.js');
const transcribeScene = require('./imageDescriber.js');
const { Storage } = require('@google-cloud/storage');
const fs = require('fs').promises;
const path = require('path');
const express = require('express');
const app = express();

const bucketName = process.env.BUCKET_ID;

const port = parseInt(process.env.PORT) || 8080;
app.listen(port, () => {
  console.log(`video describer service ready: listening on port ${port}`);
});

// entry point for the service
app.get('/', async (req, res) => {

  try {

    // download the requested video from Cloud Storage
    let videoFilename =  req.query.filename; 
    console.log("processing file: " + videoFilename);

    // download the file to locally to the Cloud Run instance
    let localFilename = await downloadVideoFile(videoFilename);

    // detect all the scenes in the video & save timestamps to an array
    let timestamps = await detectSceneChanges(localFilename);
    console.log("Detected scene changes at the following timestamps: ", timestamps);

    // create an image of each scene change
    // and save to a local directory called "output"
    await captureImages(localFilename, timestamps);

    // get an access token for the Service Account to call the Google APIs 
    let accessToken = await transcribeScene.getAccessToken();
    console.log("got an access token");

    let imageBaseName = path.parse(localFilename).name;

    // the data structure for storing the scene description and timestamp
    // e.g. an array of json objects {timestamp: 1, description: "..."}, etc.    
    let scenes = []

    // for each timestamp, send the image to Vertex AI
    console.log("getting Vertex AI description all the timestamps");
    scenes = await Promise.all(
      timestamps.map(async (timestamp) => {

        let filepath = path.join("./output", imageBaseName + "-" + timestamp + ".png");

        // get the base64 encoded image
        const encodedFile = await fs.readFile(filepath, 'base64');

        // send each screenshot to Vertex AI for description
        let description = await transcribeScene.transcribeScene(accessToken, encodedFile)

        return { timestamp: timestamp, description: description };
      }));

    console.log("finished collecting all the scenes");
    //console.log(scenes);

    return res.json(scenes);

  } catch (error) {

    //return an error
    console.log("received error: ", error);
    return res.status(500).json("an internal error occurred");
  }

});

async function downloadVideoFile(videoFilename) {
  // Creates a client
  const storage = new Storage();

  // keep same name locally
  let localFilename = videoFilename;

  const options = {
    destination: localFilename
  };

  // Download the file
  await storage.bucket(bucketName).file(videoFilename).download(options);

  console.log(
    `gs://${bucketName}/${videoFilename} downloaded locally to ${localFilename}.`
  );

  return localFilename;
}

इसके बाद, यहां दिए गए कॉन्टेंट के साथ sceneDetector.js फ़ाइल बनाएं. यह फ़ाइल, Video Intelligence API का इस्तेमाल करती है. इससे वीडियो में सीन बदलने का पता चलता है.

const fs = require('fs');
const util = require('util');
const readFile = util.promisify(fs.readFile);
const ffmpeg = require('fluent-ffmpeg');

const Video = require('@google-cloud/video-intelligence');
const client = new Video.VideoIntelligenceServiceClient();

module.exports = {
    detectSceneChanges: async function (downloadedFile) {

        // Reads a local video file and converts it to base64       
        const file = await readFile(downloadedFile);
        const inputContent = file.toString('base64');

        // setup request for shot change detection
        const videoContext = {
            speechTranscriptionConfig: {
                languageCode: 'en-US',
                enableAutomaticPunctuation: true,
            },
        };

        const request = {
            inputContent: inputContent,
            features: ['SHOT_CHANGE_DETECTION'],
        };

        // Detects camera shot changes
        const [operation] = await client.annotateVideo(request);
        console.log('Shot (scene) detection in progress...');
        const [operationResult] = await operation.promise();

        // Gets shot changes
        const shotChanges = operationResult.annotationResults[0].shotAnnotations;

        console.log("Shot (scene) changes detected: " + shotChanges.length);

        // data structure to be returned 
        let sceneChanges = [];

        // for the initial scene
        sceneChanges.push(1);

        // if only one scene, keep at 1 second
        if (shotChanges.length === 1) {
            return sceneChanges;
        }

        // get length of video
        const videoLength = await getVideoLength(downloadedFile);

        shotChanges.forEach((shot, shotIndex) => {
            if (shot.endTimeOffset === undefined) {
                shot.endTimeOffset = {};
            }
            if (shot.endTimeOffset.seconds === undefined) {
                shot.endTimeOffset.seconds = 0;
            }
            if (shot.endTimeOffset.nanos === undefined) {
                shot.endTimeOffset.nanos = 0;
            }

            // convert to a number
            let currentTimestampSecond = Number(shot.endTimeOffset.seconds);                  

            let sceneChangeTime = 0;
            // double-check no scenes were detected within the last second
            if (currentTimestampSecond + 1 > videoLength) {
                sceneChangeTime = currentTimestampSecond;                
            } else {
                // otherwise, for simplicity, just round up to the next second 
                sceneChangeTime = currentTimestampSecond + 1;
            }

            sceneChanges.push(sceneChangeTime);
        });

        return sceneChanges;
    }
}

async function getVideoLength(localFile) {
    let getLength = util.promisify(ffmpeg.ffprobe);
    let length = await getLength(localFile);

    console.log("video length: ", length.format.duration);
    return length.format.duration;
}

अब imageCapture.js नाम की फ़ाइल बनाएं और उसमें यह कॉन्टेंट शामिल करें. यह फ़ाइल, नोड ऐप्लिकेशन में ffmpeg कमांड चलाने के लिए, नोड पैकेज fluent-ffmpeg का इस्तेमाल करती है.

const ffmpeg = require('fluent-ffmpeg');
const path = require('path');
const util = require('util');


module.exports = {
    captureImages: async function (localFile, scenes) {


        let imageBaseName = path.parse(localFile).name;


        try {
            for (scene of scenes) {
                console.log("creating screenshot for scene: ", + scene);
                await createScreenshot(localFile, imageBaseName, scene);
            }


        } catch (error) {
            console.log("error gathering screenshots: ", error);
        }


        console.log("finished gathering the screenshots");
    }
}


async function createScreenshot(localFile, imageBaseName, scene) {
    return new Promise((resolve, reject) => {
        ffmpeg(localFile)
            .screenshots({
                timestamps: [scene],
                filename: `${imageBaseName}-${scene}.png`,
                folder: 'output',
                size: '320x240'
            }).on("error", () => {
                console.log("Failed to create scene for timestamp: " + scene);
                return reject('Failed to create scene for timestamp: ' + scene);
            })
            .on("end", () => {
                return resolve();
            });
    })
}

आखिर में, `imageDescriber.js`` नाम की फ़ाइल बनाएं और उसमें यह कॉन्टेंट शामिल करें. यह फ़ाइल, Vertex AI का इस्तेमाल करके हर सीन की इमेज का विज़ुअल ब्यौरा देती है.

const axios = require("axios");
const { GoogleAuth } = require('google-auth-library');

const auth = new GoogleAuth({
    scopes: 'https://www.googleapis.com/auth/cloud-platform'
});

module.exports = {
    getAccessToken: async function () {

        return await auth.getAccessToken();
    }, 

    transcribeScene: async function(token, encodedFile) {

        let projectId = await auth.getProjectId();
    
        let config = {
            headers: {
                'Authorization': 'Bearer ' + token,
                'Content-Type': 'application/json; charset=utf-8'
            }
        }

        const json = {
            "instances": [
                {
                    "image": {
                        "bytesBase64Encoded": encodedFile
                    }
                }
            ],
            "parameters": {
                "sampleCount": 1,
                "language": "en"
            }
        }

        let response = await axios.post('https://us-central1-aiplatform.googleapis.com/v1/projects/' + projectId + '/locations/us-central1/publishers/google/models/imagetext:predict', json, config);

        return response.data.predictions[0];
    }
}

Dockerfile और .dockerignore फ़ाइल बनाना

यह सेवा ffmpeg का इस्तेमाल करती है. इसलिए, आपको एक Dockerfile बनाना होगा, जो ffmpeg को इंस्टॉल करे.

Dockerfile नाम की एक फ़ाइल बनाएं. इसमें यह कॉन्टेंट शामिल करें:

# Copyright 2020 Google, LLC.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#    http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

# Use the official lightweight Node.js image.
# https://hub.docker.com/_/node
FROM node:20.10.0-slim

# Create and change to the app directory.
WORKDIR /usr/src/app

RUN apt-get update && apt-get install -y ffmpeg

# Copy application dependency manifests to the container image.
# A wildcard is used to ensure both package.json AND package-lock.json are copied.
# Copying this separately prevents re-running npm install on every code change.
COPY package*.json ./

# Install dependencies.
# If you add a package-lock.json speed your build by switching to 'npm ci'.
# RUN npm ci --only=production
RUN npm install --production

# Copy local code to the container image.
COPY . .

# Run the web service on container startup.
CMD [ "npm", "start" ]

इसके अलावा, कुछ फ़ाइलों को कंटेनर में शामिल न करने के लिए, .dockerignore नाम की फ़ाइल बनाएं.

Dockerfile
.dockerignore
node_modules
npm-debug.log

6. सेवा खाता बनाना

Cloud Run सेवा के लिए एक सेवा खाता बनाया जाएगा. इसका इस्तेमाल Cloud Storage, Vertex AI, और Video Intelligence API को ऐक्सेस करने के लिए किया जाएगा.

SERVICE_ACCOUNT="cloud-run-video-description"
SERVICE_ACCOUNT_ADDRESS=$SERVICE_ACCOUNT@$PROJECT_ID.iam.gserviceaccount.com

gcloud iam service-accounts create $SERVICE_ACCOUNT \
  --display-name="Cloud Run Video Scene Image Describer service account"
 
# to view & download storage bucket objects
gcloud projects add-iam-policy-binding $PROJECT_ID \
  --member serviceAccount:$SERVICE_ACCOUNT_ADDRESS \
  --role=roles/storage.objectViewer

# to call the Vertex AI imagetext model
gcloud projects add-iam-policy-binding $PROJECT_ID \
  --member serviceAccount:$SERVICE_ACCOUNT_ADDRESS \
  --role=roles/aiplatform.user

7. Cloud Run सेवा को डिप्लॉय करना

अब सोर्स पर आधारित डिप्लॉयमेंट का इस्तेमाल करके, Cloud Run सेवा को अपने-आप कंटेनर में बदला जा सकता है.

ध्यान दें: Cloud Run सेवा के लिए, डिफ़ॉल्ट प्रोसेसिंग का समय 60 सेकंड होता है. इस कोडलैब में पांच मिनट का टाइमआउट इस्तेमाल किया गया है, क्योंकि सुझाया गया टेस्ट वीडियो दो मिनट का है. अगर आपको ज़्यादा अवधि वाला वीडियो इस्तेमाल करना है, तो आपको समय में बदलाव करना पड़ सकता है.

gcloud run deploy $SERVICE_NAME \
  --region=$REGION \
  --set-env-vars BUCKET_ID=$BUCKET_ID \
  --no-allow-unauthenticated \
  --service-account $SERVICE_ACCOUNT_ADDRESS \
  --timeout=5m \
  --source=.

डप्लॉय करने के बाद, सेवा के यूआरएल को एनवायरमेंट वैरिएबल में सेव करें.

SERVICE_URL=$(gcloud run services describe $SERVICE_NAME --platform managed --region $REGION --format 'value(status.url)')

8. Cloud Run सेवा को कॉल करना

अब Cloud Storage में अपलोड किए गए वीडियो का नाम देकर, अपनी सेवा को कॉल किया जा सकता है.

curl -X GET -H "Authorization: Bearer $(gcloud auth print-identity-token)" ${SERVICE_URL}?filename=${FILENAME}

आपके नतीजे, यहाँ दिए गए उदाहरण के आउटपुट से मिलते-जुलते होने चाहिए:

[{"timestamp":1,"description":"an aerial view of a city with a bridge in the background"},{"timestamp":7,"description":"a man in a blue shirt sits in front of shelves of donuts"},{"timestamp":11,"description":"a black and white photo of people working in a bakery"},{"timestamp":12,"description":"a black and white photo of a man and woman working in a bakery"}]

9. बधाई हो!

कोडलैब पूरा करने के लिए बधाई!

हमारा सुझाव है कि आप Video Intelligence API, Cloud Run, और Vertex AI की विज़ुअल कैप्शनिंग की सुविधा के बारे में दस्तावेज़ पढ़ें.

हमने क्या-क्या बताया

  • तीसरे पक्ष की बाइनरी इंस्टॉल करने के लिए, Dockerfile का इस्तेमाल करके कंटेनर इमेज बनाने का तरीका
  • Cloud Run सेवा के लिए सेवा खाता बनाकर, कम से कम विशेषाधिकार के सिद्धांत का पालन कैसे करें, ताकि वह अन्य Google Cloud सेवाओं को कॉल कर सके
  • Cloud Run सेवा से Video Intelligence क्लाइंट लाइब्रेरी का इस्तेमाल करने का तरीका
  • Vertex AI से हर सीन की विज़ुअल जानकारी पाने के लिए, Google API को कॉल करने का तरीका

10. व्यवस्थित करें

अनजाने में लगने वाले शुल्क से बचने के लिए, (उदाहरण के लिए, अगर इस Cloud Run सेवा को मुफ़्त टियर में Cloud Run के लिए महीने के तय किए गए कॉल की संख्या से ज़्यादा बार अनजाने में कॉल किया जाता है), तो Cloud Run सेवा को मिटा दें या चरण 2 में बनाए गए प्रोजेक्ट को मिटा दें.

Cloud Run सेवा को मिटाने के लिए, Cloud Run Cloud Console पर https://console.cloud.google.com/run/ पर जाएं. इसके बाद, video-describer फ़ंक्शन (या अगर आपने किसी दूसरे नाम का इस्तेमाल किया है, तो $SERVICE_NAME) को मिटाएं.

अगर आपको पूरा प्रोजेक्ट मिटाना है, तो https://console.cloud.google.com/cloud-resource-manager पर जाएं. इसके बाद, दूसरे चरण में बनाया गया प्रोजेक्ट चुनें और मिटाएं को चुनें. प्रोजेक्ट मिटाने पर, आपको Cloud SDK में प्रोजेक्ट बदलने होंगे. gcloud projects list कमांड चलाकर, सभी उपलब्ध प्रोजेक्ट की सूची देखी जा सकती है.