Riconoscimento dei loghi

L'API Video Intelligence è in grado di rilevare, monitorare e riconoscere la presenza di oltre 100.000 brand e loghi nei contenuti video.

Questa pagina descrive come riconoscere un logo in un video utilizzando l'API Video Intelligence.

Annotare un video in Cloud Storage

Il seguente esempio di codice mostra come rilevare i loghi in un video in Cloud Storage.

REST

Invia la richiesta di elaborazione

Per eseguire l'annotazione su un file video locale, codifica i contenuti del file video in Base64. Includi i contenuti codificati in Base64 nel campo inputContent della richiesta. Per informazioni su come codificare in Base64 i contenuti di un file video, consulta Codifica Base64.

Di seguito viene mostrato come inviare una richiesta POST al metodo videos:annotate. L'esempio utilizza il token di accesso per un account di servizio configurato per il progetto utilizzando Google Cloud CLI. Per istruzioni su come installare Google Cloud CLI, configurare un progetto con un account di servizio e ottenere un token di accesso, consulta la guida rapida di Video Intelligence.

Prima di utilizzare i dati della richiesta, apporta le seguenti sostituzioni:

  • INPUT_URI: un bucket Cloud Storage contenente il file da annotare, incluso il nome del file. Deve iniziare con gs://.
    Ad esempio:
    "inputUri": "gs://cloud-videointelligence-demo/assistant.mp4",
  • PROJECT_NUMBER: l'identificatore numerico del tuo progetto Google Cloud

Metodo HTTP e URL: 

POST https://videointelligence.googleapis.com/v1/videos:annotate

Corpo JSON della richiesta: 

{
    "inputUri":"INPUT_URI",
    "features": ["LOGO_RECOGNITION"]
}

Per inviare la richiesta, espandi una di queste opzioni:

Dovresti ricevere una risposta JSON simile alla seguente:

{
  "name": "projects/PROJECT_NUMBER/locations/LOCATION_ID/operations/OPERATION_ID"
}

Se la risposta è positiva, l'API Video Intelligence restituisce il name per l'operazione. Di seguito è riportato un esempio di questa risposta, in cui: project-number è il numero del progetto e operation-id è l'ID dell'operazione di lunga durata creata per la richiesta.

  • PROJECT_NUMBER: il numero del progetto
  • LOCATION_ID: la regione Cloud in cui deve avvenire l'annotazione. Le regioni cloud supportate sono: us-east1, us-west1, europe-west1, asia-east1. Se non viene specificata alcuna regione, verrà determinata una regione in base alla posizione del file video.
  • OPERATION_ID: l'ID dell'operazione a lunga esecuzione creata per la richiesta e fornito nella risposta quando hai avviato l'operazione, ad esempio 12345...

Visualizza i risultati

Per ottenere i risultati della richiesta, invia una richiesta GET utilizzando il nome dell'operazione restituito dalla chiamata a videos:annotate, come mostrato nell'esempio seguente.

Prima di utilizzare i dati della richiesta, apporta le seguenti sostituzioni:

  • OPERATION_NAME: il nome dell'operazione come fornito dall'API Video Intelligence. Il nome dell'operazione ha il formato projects/PROJECT_NUMBER/locations/LOCATION_ID/operations/OPERATION_ID
  • PROJECT_NUMBER: l'identificatore numerico del tuo progetto Google Cloud

Metodo HTTP e URL: 

GET https://videointelligence.googleapis.com/v1/OPERATION_NAME

Per inviare la richiesta, espandi una di queste opzioni:

Dovresti ricevere una risposta JSON simile alla seguente:

Scaricare i risultati dell'annotazione

Copia l'annotazione dal bucket di origine a quello di destinazione: (vedi Copiare file e oggetti)

gcloud storage cp gcs_uri gs://my-bucket

Nota: se l'URI GCS di output viene fornito dall'utente, l'annotazione viene archiviata in quell'URI GCS.

Go

Per autenticarti a Video Intelligence, configura le Credenziali predefinite dell'applicazione. Per ulteriori informazioni, consulta Configurare l'autenticazione per un ambiente di sviluppo locale. 

import (
	"context"
	"fmt"
	"io"
	"time"

	video "cloud.google.com/go/videointelligence/apiv1"
	videopb "cloud.google.com/go/videointelligence/apiv1/videointelligencepb"
	"github.com/golang/protobuf/ptypes"
)

// logoDetectionGCS analyzes a video and extracts logos with their bounding boxes.
func logoDetectionGCS(w io.Writer, gcsURI string) error {
	// gcsURI := "gs://cloud-samples-data/video/googlework_tiny.mp4"

	ctx := context.Background()

	// Creates a client.
	client, err := video.NewClient(ctx)
	if err != nil {
		return fmt.Errorf("video.NewClient: %w", err)
	}
	defer client.Close()

	ctx, cancel := context.WithTimeout(ctx, time.Second*180)
	defer cancel()

	op, err := client.AnnotateVideo(ctx, &videopb.AnnotateVideoRequest{
		InputUri: gcsURI,
		Features: []videopb.Feature{
			videopb.Feature_LOGO_RECOGNITION,
		},
	})
	if err != nil {
		return fmt.Errorf("AnnotateVideo: %w", err)
	}

	resp, err := op.Wait(ctx)
	if err != nil {
		return fmt.Errorf("Wait: %w", err)
	}

	// Only one video was processed, so get the first result.
	result := resp.GetAnnotationResults()[0]

	// Annotations for list of logos detected, tracked and recognized in video.
	for _, annotation := range result.LogoRecognitionAnnotations {
		fmt.Fprintf(w, "Description: %q\n", annotation.Entity.GetDescription())
		// Opaque entity ID. Some IDs may be available in Google Knowledge
		// Graph Search API (https://developers.google.com/knowledge-graph/).
		if len(annotation.Entity.EntityId) > 0 {
			fmt.Fprintf(w, "\tEntity ID: %q\n", annotation.Entity.GetEntityId())
		}

		// All logo tracks where the recognized logo appears. Each track
		// corresponds to one logo instance appearing in consecutive frames.
		for _, track := range annotation.Tracks {
			// Video segment of a track.
			segment := track.GetSegment()
			start, _ := ptypes.Duration(segment.GetStartTimeOffset())
			end, _ := ptypes.Duration(segment.GetEndTimeOffset())
			fmt.Fprintf(w, "\tSegment: %v to %v\n", start, end)
			fmt.Fprintf(w, "\tConfidence: %f\n", track.GetConfidence())

			// The object with timestamp and attributes per frame in the track.
			for _, timestampedObject := range track.TimestampedObjects {
				// Normalized Bounding box in a frame, where the object is
				// located.
				box := timestampedObject.GetNormalizedBoundingBox()
				fmt.Fprintf(w, "\tBounding box position:\n")
				fmt.Fprintf(w, "\t\tleft  : %f\n", box.GetLeft())
				fmt.Fprintf(w, "\t\ttop   : %f\n", box.GetTop())
				fmt.Fprintf(w, "\t\tright : %f\n", box.GetRight())
				fmt.Fprintf(w, "\t\tbottom: %f\n", box.GetBottom())

				// Optional. The attributes of the object in the bounding box.
				for _, attribute := range timestampedObject.Attributes {
					fmt.Fprintf(w, "\t\t\tName: %q\n", attribute.GetName())
					fmt.Fprintf(w, "\t\t\tConfidence: %f\n", attribute.GetConfidence())
					fmt.Fprintf(w, "\t\t\tValue: %q\n", attribute.GetValue())
				}
			}

			// Optional. Attributes in the track level.
			for _, trackAttribute := range track.Attributes {
				fmt.Fprintf(w, "\t\tName: %q\n", trackAttribute.GetName())
				fmt.Fprintf(w, "\t\tConfidence: %f\n", trackAttribute.GetConfidence())
				fmt.Fprintf(w, "\t\tValue: %q\n", trackAttribute.GetValue())
			}
		}

		// All video segments where the recognized logo appears. There might be
		// multiple instances of the same logo class appearing in one VideoSegment.
		for _, segment := range annotation.Segments {
			start, _ := ptypes.Duration(segment.GetStartTimeOffset())
			end, _ := ptypes.Duration(segment.GetEndTimeOffset())
			fmt.Fprintf(w, "\tSegment: %v to %v\n", start, end)
		}
	}

	return nil
}

Java

Per autenticarti a Video Intelligence, configura le Credenziali predefinite dell'applicazione. Per ulteriori informazioni, consulta Configurare l'autenticazione per un ambiente di sviluppo locale. 


import com.google.api.gax.longrunning.OperationFuture;
import com.google.cloud.videointelligence.v1.AnnotateVideoProgress;
import com.google.cloud.videointelligence.v1.AnnotateVideoRequest;
import com.google.cloud.videointelligence.v1.AnnotateVideoResponse;
import com.google.cloud.videointelligence.v1.DetectedAttribute;
import com.google.cloud.videointelligence.v1.Entity;
import com.google.cloud.videointelligence.v1.Feature;
import com.google.cloud.videointelligence.v1.LogoRecognitionAnnotation;
import com.google.cloud.videointelligence.v1.NormalizedBoundingBox;
import com.google.cloud.videointelligence.v1.TimestampedObject;
import com.google.cloud.videointelligence.v1.Track;
import com.google.cloud.videointelligence.v1.VideoAnnotationResults;
import com.google.cloud.videointelligence.v1.VideoIntelligenceServiceClient;
import com.google.cloud.videointelligence.v1.VideoSegment;
import com.google.protobuf.Duration;
import java.io.IOException;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.TimeoutException;

public class LogoDetectionGcs {

  public static void detectLogoGcs() throws Exception {
    // TODO(developer): Replace these variables before running the sample.
    String gcsUri = "gs://YOUR_BUCKET_ID/path/to/your/video.mp4";
    detectLogoGcs(gcsUri);
  }

  public static void detectLogoGcs(String inputUri)
      throws IOException, ExecutionException, InterruptedException, TimeoutException {
    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (VideoIntelligenceServiceClient client = VideoIntelligenceServiceClient.create()) {
      // Create the request
      AnnotateVideoRequest request =
          AnnotateVideoRequest.newBuilder()
              .setInputUri(inputUri)
              .addFeatures(Feature.LOGO_RECOGNITION)
              .build();

      // asynchronously perform object tracking on videos
      OperationFuture<AnnotateVideoResponse, AnnotateVideoProgress> future =
          client.annotateVideoAsync(request);

      System.out.println("Waiting for operation to complete...");
      // The first result is retrieved because a single video was processed.
      AnnotateVideoResponse response = future.get(600, TimeUnit.SECONDS);
      VideoAnnotationResults annotationResult = response.getAnnotationResults(0);

      // Annotations for list of logos detected, tracked and recognized in video.
      for (LogoRecognitionAnnotation logoRecognitionAnnotation :
          annotationResult.getLogoRecognitionAnnotationsList()) {
        Entity entity = logoRecognitionAnnotation.getEntity();
        // Opaque entity ID. Some IDs may be available in
        // [Google Knowledge Graph Search API](https://developers.google.com/knowledge-graph/).
        System.out.printf("Entity Id : %s\n", entity.getEntityId());
        System.out.printf("Description : %s\n", entity.getDescription());
        // All logo tracks where the recognized logo appears. Each track corresponds to one logo
        // instance appearing in consecutive frames.
        for (Track track : logoRecognitionAnnotation.getTracksList()) {

          // Video segment of a track.
          Duration startTimeOffset = track.getSegment().getStartTimeOffset();
          System.out.printf(
              "\n\tStart Time Offset: %s.%s\n",
              startTimeOffset.getSeconds(), startTimeOffset.getNanos());
          Duration endTimeOffset = track.getSegment().getEndTimeOffset();
          System.out.printf(
              "\tEnd Time Offset: %s.%s\n", endTimeOffset.getSeconds(), endTimeOffset.getNanos());
          System.out.printf("\tConfidence: %s\n", track.getConfidence());

          // The object with timestamp and attributes per frame in the track.
          for (TimestampedObject timestampedObject : track.getTimestampedObjectsList()) {

            // Normalized Bounding box in a frame, where the object is located.
            NormalizedBoundingBox normalizedBoundingBox =
                timestampedObject.getNormalizedBoundingBox();
            System.out.printf("\n\t\tLeft: %s\n", normalizedBoundingBox.getLeft());
            System.out.printf("\t\tTop: %s\n", normalizedBoundingBox.getTop());
            System.out.printf("\t\tRight: %s\n", normalizedBoundingBox.getRight());
            System.out.printf("\t\tBottom: %s\n", normalizedBoundingBox.getBottom());

            // Optional. The attributes of the object in the bounding box.
            for (DetectedAttribute attribute : timestampedObject.getAttributesList()) {
              System.out.printf("\n\t\t\tName: %s\n", attribute.getName());
              System.out.printf("\t\t\tConfidence: %s\n", attribute.getConfidence());
              System.out.printf("\t\t\tValue: %s\n", attribute.getValue());
            }
          }

          // Optional. Attributes in the track level.
          for (DetectedAttribute trackAttribute : track.getAttributesList()) {
            System.out.printf("\n\t\tName : %s\n", trackAttribute.getName());
            System.out.printf("\t\tConfidence : %s\n", trackAttribute.getConfidence());
            System.out.printf("\t\tValue : %s\n", trackAttribute.getValue());
          }
        }

        // All video segments where the recognized logo appears. There might be multiple instances
        // of the same logo class appearing in one VideoSegment.
        for (VideoSegment segment : logoRecognitionAnnotation.getSegmentsList()) {
          System.out.printf(
              "\n\tStart Time Offset : %s.%s\n",
              segment.getStartTimeOffset().getSeconds(), segment.getStartTimeOffset().getNanos());
          System.out.printf(
              "\tEnd Time Offset : %s.%s\n",
              segment.getEndTimeOffset().getSeconds(), segment.getEndTimeOffset().getNanos());
        }
      }
    }
  }
}

Node.js

Per autenticarti a Video Intelligence, configura le Credenziali predefinite dell'applicazione. Per ulteriori informazioni, consulta Configurare l'autenticazione per un ambiente di sviluppo locale. 

/**
 * TODO(developer): Uncomment these variables before running the sample.
 */
// const inputUri = 'gs://cloud-samples-data/video/googlework_short.mp4';

// Imports the Google Cloud client libraries
const Video = require('@google-cloud/video-intelligence');

// Instantiates a client
const client = new Video.VideoIntelligenceServiceClient();

// Performs asynchronous video annotation for logo recognition on a file hosted in GCS.
async function detectLogoGcs() {
  // Build the request with the input uri and logo recognition feature.
  const request = {
    inputUri: inputUri,
    features: ['LOGO_RECOGNITION'],
  };

  // Make the asynchronous request
  const [operation] = await client.annotateVideo(request);

  // Wait for the results
  const [response] = await operation.promise();

  // Get the first response, since we sent only one video.
  const annotationResult = response.annotationResults[0];
  for (const logoRecognitionAnnotation of annotationResult.logoRecognitionAnnotations) {
    const entity = logoRecognitionAnnotation.entity;
    // Opaque entity ID. Some IDs may be available in
    // [Google Knowledge Graph Search API](https://developers.google.com/knowledge-graph/).
    console.log(`Entity Id: ${entity.entityId}`);
    console.log(`Description: ${entity.description}`);

    // All logo tracks where the recognized logo appears.
    // Each track corresponds to one logo instance appearing in consecutive frames.
    for (const track of logoRecognitionAnnotation.tracks) {
      console.log(
        `\n\tStart Time Offset: ${track.segment.startTimeOffset.seconds}.${track.segment.startTimeOffset.nanos}`
      );
      console.log(
        `\tEnd Time Offset: ${track.segment.endTimeOffset.seconds}.${track.segment.endTimeOffset.nanos}`
      );
      console.log(`\tConfidence: ${track.confidence}`);

      // The object with timestamp and attributes per frame in the track.
      for (const timestampedObject of track.timestampedObjects) {
        // Normalized Bounding box in a frame, where the object is located.
        const normalizedBoundingBox = timestampedObject.normalizedBoundingBox;
        console.log(`\n\t\tLeft: ${normalizedBoundingBox.left}`);
        console.log(`\t\tTop: ${normalizedBoundingBox.top}`);
        console.log(`\t\tRight: ${normalizedBoundingBox.right}`);
        console.log(`\t\tBottom: ${normalizedBoundingBox.bottom}`);
        // Optional. The attributes of the object in the bounding box.
        for (const attribute of timestampedObject.attributes) {
          console.log(`\n\t\t\tName: ${attribute.name}`);
          console.log(`\t\t\tConfidence: ${attribute.confidence}`);
          console.log(`\t\t\tValue: ${attribute.value}`);
        }
      }

      // Optional. Attributes in the track level.
      for (const trackAttribute of track.attributes) {
        console.log(`\n\t\tName: ${trackAttribute.name}`);
        console.log(`\t\tConfidence: ${trackAttribute.confidence}`);
        console.log(`\t\tValue: ${trackAttribute.value}`);
      }
    }

    // All video segments where the recognized logo appears.
    // There might be multiple instances of the same logo class appearing in one VideoSegment.
    for (const segment of logoRecognitionAnnotation.segments) {
      console.log(
        `\n\tStart Time Offset: ${segment.startTimeOffset.seconds}.${segment.startTimeOffset.nanos}`
      );
      console.log(
        `\tEnd Time Offset: ${segment.endTimeOffset.seconds}.${segment.endTimeOffset.nanos}`
      );
    }
  }
}

detectLogoGcs();

Python

Per autenticarti a Video Intelligence, configura le Credenziali predefinite dell'applicazione. Per ulteriori informazioni, consulta Configurare l'autenticazione per un ambiente di sviluppo locale. 


from google.cloud import videointelligence


def detect_logo_gcs(input_uri="gs://YOUR_BUCKET_ID/path/to/your/file.mp4"):
    client = videointelligence.VideoIntelligenceServiceClient()

    features = [videointelligence.Feature.LOGO_RECOGNITION]

    operation = client.annotate_video(
        request={"features": features, "input_uri": input_uri}
    )

    print("Waiting for operation to complete...")
    response = operation.result()

    # Get the first response, since we sent only one video.
    annotation_result = response.annotation_results[0]

    # Annotations for list of logos detected, tracked and recognized in video.
    for logo_recognition_annotation in annotation_result.logo_recognition_annotations:
        entity = logo_recognition_annotation.entity

        # Opaque entity ID. Some IDs may be available in [Google Knowledge Graph
        # Search API](https://developers.google.com/knowledge-graph/).
        print("Entity Id : {}".format(entity.entity_id))

        print("Description : {}".format(entity.description))

        # All logo tracks where the recognized logo appears. Each track corresponds
        # to one logo instance appearing in consecutive frames.
        for track in logo_recognition_annotation.tracks:
            # Video segment of a track.
            print(
                "\n\tStart Time Offset : {}.{}".format(
                    track.segment.start_time_offset.seconds,
                    track.segment.start_time_offset.microseconds * 1000,
                )
            )
            print(
                "\tEnd Time Offset : {}.{}".format(
                    track.segment.end_time_offset.seconds,
                    track.segment.end_time_offset.microseconds * 1000,
                )
            )
            print("\tConfidence : {}".format(track.confidence))

            # The object with timestamp and attributes per frame in the track.
            for timestamped_object in track.timestamped_objects:
                # Normalized Bounding box in a frame, where the object is located.
                normalized_bounding_box = timestamped_object.normalized_bounding_box
                print("\n\t\tLeft : {}".format(normalized_bounding_box.left))
                print("\t\tTop : {}".format(normalized_bounding_box.top))
                print("\t\tRight : {}".format(normalized_bounding_box.right))
                print("\t\tBottom : {}".format(normalized_bounding_box.bottom))

                # Optional. The attributes of the object in the bounding box.
                for attribute in timestamped_object.attributes:
                    print("\n\t\t\tName : {}".format(attribute.name))
                    print("\t\t\tConfidence : {}".format(attribute.confidence))
                    print("\t\t\tValue : {}".format(attribute.value))

            # Optional. Attributes in the track level.
            for track_attribute in track.attributes:
                print("\n\t\tName : {}".format(track_attribute.name))
                print("\t\tConfidence : {}".format(track_attribute.confidence))
                print("\t\tValue : {}".format(track_attribute.value))

        # All video segments where the recognized logo appears. There might be
        # multiple instances of the same logo class appearing in one VideoSegment.
        for segment in logo_recognition_annotation.segments:
            print(
                "\n\tStart Time Offset : {}.{}".format(
                    segment.start_time_offset.seconds,
                    segment.start_time_offset.microseconds * 1000,
                )
            )
            print(
                "\tEnd Time Offset : {}.{}".format(
                    segment.end_time_offset.seconds,
                    segment.end_time_offset.microseconds * 1000,
                )
            )

Linguaggi aggiuntivi

C#: segui le istruzioni di configurazione per C# nella pagina delle librerie client e poi consulta la documentazione di riferimento di Video Intelligence per .NET.

PHP: segui le istruzioni di configurazione di PHP riportate nella pagina delle librerie client e consulta la documentazione di riferimento di Video Intelligence per PHP.

Ruby: segui le istruzioni di configurazione di Ruby nella pagina delle librerie client e poi consulta la documentazione di riferimento di Video Intelligence per Ruby.

Annotare un video locale

Il seguente esempio di codice mostra come rilevare i loghi in un file video locale.

REST

Inviare una richiesta di annotazione video

Per eseguire l'annotazione su un file video locale, assicurati di codificare in base64 i contenuti del file video. Includi i contenuti codificati in base64 nel campo inputContent della richiesta. Per informazioni su come codificare in Base64 i contenuti di un file video, consulta Codifica Base64.

Di seguito è riportato un esempio di come inviare una richiesta POST al metodo videos:annotate. L'esempio utilizza il token di accesso per un account di servizio configurato per il progetto utilizzando Google Cloud CLI. Per istruzioni su come installare Google Cloud CLI, configurare un progetto con un account di servizio e ottenere un token di accesso, consulta la guida introduttiva all'API Video Intelligence.

Prima di utilizzare i dati della richiesta, apporta le seguenti sostituzioni:

  • "inputContent": BASE64_ENCODED_CONTENT
    Ad esempio:
    "UklGRg41AwBBVkkgTElTVAwBAABoZHJsYXZpaDgAAAA1ggAAxPMBAAAAAAAQCAA..."
  • LANGUAGE_CODE: [Facoltativo] consulta le lingue supportate
  • PROJECT_NUMBER: l'identificatore numerico del tuo progetto Google Cloud

Metodo HTTP e URL: 

POST https://videointelligence.googleapis.com/v1/videos:annotate

Corpo JSON della richiesta: 

{
  "inputContent": "BASE64_ENCODED_CONTENT",
  "features": ["LOGO_RECOGNITION"],
  "videoContext": {
  }
}

Per inviare la richiesta, espandi una di queste opzioni:

Dovresti ricevere una risposta JSON simile alla seguente:

{
  "name": "projects/PROJECT_NUMBER/locations/LOCATION_ID/operations/OPERATION_ID"
}

Se la risposta è positiva, l'API Video Intelligence restituisce il name per l'operazione. Di seguito è riportato un esempio di questa risposta, dove project-number è il nome del progetto e operation-id è l'ID dell'operazione in esecuzione prolungata creata per la richiesta.

  • OPERATION_ID: fornito nella risposta quando hai avviato l'operazione, ad esempio 12345...

Recuperare i risultati dell'annotazione

Per recuperare il risultato dell'operazione, effettua una richiesta GET, utilizzando il nome dell'operazione restituito dalla chiamata a videos:annotate, come mostrato nell'esempio seguente.

Prima di utilizzare i dati della richiesta, apporta le seguenti sostituzioni:

  • PROJECT_NUMBER: l'identificatore numerico del tuo progetto Google Cloud

Metodo HTTP e URL: 

GET https://videointelligence.googleapis.com/v1/OPERATION_NAME

Per inviare la richiesta, espandi una di queste opzioni:

Dovresti ricevere una risposta JSON simile alla seguente:

Le annotazioni per il rilevamento del testo vengono restituite come elenco textAnnotations. Nota: il campo done viene restituito solo quando il relativo valore è True. Non è incluso nelle risposte per le quali l'operazione non è stata completata.

Go

Per autenticarti a Video Intelligence, configura le Credenziali predefinite dell'applicazione. Per ulteriori informazioni, consulta Configurare l'autenticazione per un ambiente di sviluppo locale. 

import (
	"context"
	"fmt"
	"io"
	"os"
	"time"

	video "cloud.google.com/go/videointelligence/apiv1"
	videopb "cloud.google.com/go/videointelligence/apiv1/videointelligencepb"
	"github.com/golang/protobuf/ptypes"
)

// logoDetection analyzes a video and extracts logos with their bounding boxes.
func logoDetection(w io.Writer, filename string) error {
	// filename := "../testdata/googlework_short.mp4"

	ctx := context.Background()

	// Creates a client.
	client, err := video.NewClient(ctx)
	if err != nil {
		return fmt.Errorf("video.NewClient: %w", err)
	}
	defer client.Close()

	ctx, cancel := context.WithTimeout(ctx, time.Second*180)
	defer cancel()

	fileBytes, err := os.ReadFile(filename)
	if err != nil {
		return fmt.Errorf("os.ReadFile: %w", err)
	}

	op, err := client.AnnotateVideo(ctx, &videopb.AnnotateVideoRequest{
		InputContent: fileBytes,
		Features: []videopb.Feature{
			videopb.Feature_LOGO_RECOGNITION,
		},
	})
	if err != nil {
		return fmt.Errorf("AnnotateVideo: %w", err)
	}

	resp, err := op.Wait(ctx)
	if err != nil {
		return fmt.Errorf("Wait: %w", err)
	}

	// Only one video was processed, so get the first result.
	result := resp.GetAnnotationResults()[0]

	// Annotations for list of logos detected, tracked and recognized in video.
	for _, annotation := range result.LogoRecognitionAnnotations {
		fmt.Fprintf(w, "Description: %q\n", annotation.Entity.GetDescription())
		// Opaque entity ID. Some IDs may be available in Google Knowledge
		// Graph Search API (https://developers.google.com/knowledge-graph/).
		if len(annotation.Entity.EntityId) > 0 {
			fmt.Fprintf(w, "\tEntity ID: %q\n", annotation.Entity.GetEntityId())
		}

		// All logo tracks where the recognized logo appears. Each track
		// corresponds to one logo instance appearing in consecutive frames.
		for _, track := range annotation.Tracks {
			// Video segment of a track.
			segment := track.GetSegment()
			start, _ := ptypes.Duration(segment.GetStartTimeOffset())
			end, _ := ptypes.Duration(segment.GetEndTimeOffset())
			fmt.Fprintf(w, "\tSegment: %v to %v\n", start, end)
			fmt.Fprintf(w, "\tConfidence: %f\n", track.GetConfidence())

			// The object with timestamp and attributes per frame in the track.
			for _, timestampedObject := range track.TimestampedObjects {
				// Normalized Bounding box in a frame, where the object is
				// located.
				box := timestampedObject.GetNormalizedBoundingBox()
				fmt.Fprintf(w, "\tBounding box position:\n")
				fmt.Fprintf(w, "\t\tleft  : %f\n", box.GetLeft())
				fmt.Fprintf(w, "\t\ttop   : %f\n", box.GetTop())
				fmt.Fprintf(w, "\t\tright : %f\n", box.GetRight())
				fmt.Fprintf(w, "\t\tbottom: %f\n", box.GetBottom())

				// Optional. The attributes of the object in the bounding box.
				for _, attribute := range timestampedObject.Attributes {
					fmt.Fprintf(w, "\t\t\tName: %q\n", attribute.GetName())
					fmt.Fprintf(w, "\t\t\tConfidence: %f\n", attribute.GetConfidence())
					fmt.Fprintf(w, "\t\t\tValue: %q\n", attribute.GetValue())
				}
			}

			// Optional. Attributes in the track level.
			for _, trackAttribute := range track.Attributes {
				fmt.Fprintf(w, "\t\tName: %q\n", trackAttribute.GetName())
				fmt.Fprintf(w, "\t\tConfidence: %f\n", trackAttribute.GetConfidence())
				fmt.Fprintf(w, "\t\tValue: %q\n", trackAttribute.GetValue())
			}
		}

		// All video segments where the recognized logo appears. There might be
		// multiple instances of the same logo class appearing in one VideoSegment.
		for _, segment := range annotation.Segments {
			start, _ := ptypes.Duration(segment.GetStartTimeOffset())
			end, _ := ptypes.Duration(segment.GetEndTimeOffset())
			fmt.Fprintf(w, "\tSegment: %v to %v\n", start, end)
		}
	}

	return nil
}

Java

Per autenticarti a Video Intelligence, configura le Credenziali predefinite dell'applicazione. Per ulteriori informazioni, consulta Configurare l'autenticazione per un ambiente di sviluppo locale. 


import com.google.api.gax.longrunning.OperationFuture;
import com.google.cloud.videointelligence.v1.AnnotateVideoProgress;
import com.google.cloud.videointelligence.v1.AnnotateVideoRequest;
import com.google.cloud.videointelligence.v1.AnnotateVideoResponse;
import com.google.cloud.videointelligence.v1.DetectedAttribute;
import com.google.cloud.videointelligence.v1.Entity;
import com.google.cloud.videointelligence.v1.Feature;
import com.google.cloud.videointelligence.v1.LogoRecognitionAnnotation;
import com.google.cloud.videointelligence.v1.NormalizedBoundingBox;
import com.google.cloud.videointelligence.v1.TimestampedObject;
import com.google.cloud.videointelligence.v1.Track;
import com.google.cloud.videointelligence.v1.VideoAnnotationResults;
import com.google.cloud.videointelligence.v1.VideoIntelligenceServiceClient;
import com.google.cloud.videointelligence.v1.VideoSegment;
import com.google.protobuf.ByteString;
import com.google.protobuf.Duration;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.TimeoutException;

public class LogoDetection {

  public static void detectLogo() throws Exception {
    // TODO(developer): Replace these variables before running the sample.
    String localFilePath = "path/to/your/video.mp4";
    detectLogo(localFilePath);
  }

  public static void detectLogo(String filePath)
      throws IOException, ExecutionException, InterruptedException, TimeoutException {
    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (VideoIntelligenceServiceClient client = VideoIntelligenceServiceClient.create()) {
      // Read file
      Path path = Paths.get(filePath);
      byte[] data = Files.readAllBytes(path);
      // Create the request
      AnnotateVideoRequest request =
          AnnotateVideoRequest.newBuilder()
              .setInputContent(ByteString.copyFrom(data))
              .addFeatures(Feature.LOGO_RECOGNITION)
              .build();

      // asynchronously perform object tracking on videos
      OperationFuture<AnnotateVideoResponse, AnnotateVideoProgress> future =
          client.annotateVideoAsync(request);

      System.out.println("Waiting for operation to complete...");
      // The first result is retrieved because a single video was processed.
      AnnotateVideoResponse response = future.get(300, TimeUnit.SECONDS);
      VideoAnnotationResults annotationResult = response.getAnnotationResults(0);

      // Annotations for list of logos detected, tracked and recognized in video.
      for (LogoRecognitionAnnotation logoRecognitionAnnotation :
          annotationResult.getLogoRecognitionAnnotationsList()) {
        Entity entity = logoRecognitionAnnotation.getEntity();
        // Opaque entity ID. Some IDs may be available in
        // [Google Knowledge Graph Search API](https://developers.google.com/knowledge-graph/).
        System.out.printf("Entity Id : %s\n", entity.getEntityId());
        System.out.printf("Description : %s\n", entity.getDescription());
        // All logo tracks where the recognized logo appears. Each track corresponds to one logo
        // instance appearing in consecutive frames.
        for (Track track : logoRecognitionAnnotation.getTracksList()) {

          // Video segment of a track.
          Duration startTimeOffset = track.getSegment().getStartTimeOffset();
          System.out.printf(
              "\n\tStart Time Offset: %s.%s\n",
              startTimeOffset.getSeconds(), startTimeOffset.getNanos());
          Duration endTimeOffset = track.getSegment().getEndTimeOffset();
          System.out.printf(
              "\tEnd Time Offset: %s.%s\n", endTimeOffset.getSeconds(), endTimeOffset.getNanos());
          System.out.printf("\tConfidence: %s\n", track.getConfidence());

          // The object with timestamp and attributes per frame in the track.
          for (TimestampedObject timestampedObject : track.getTimestampedObjectsList()) {

            // Normalized Bounding box in a frame, where the object is located.
            NormalizedBoundingBox normalizedBoundingBox =
                timestampedObject.getNormalizedBoundingBox();
            System.out.printf("\n\t\tLeft: %s\n", normalizedBoundingBox.getLeft());
            System.out.printf("\t\tTop: %s\n", normalizedBoundingBox.getTop());
            System.out.printf("\t\tRight: %s\n", normalizedBoundingBox.getRight());
            System.out.printf("\t\tBottom: %s\n", normalizedBoundingBox.getBottom());

            // Optional. The attributes of the object in the bounding box.
            for (DetectedAttribute attribute : timestampedObject.getAttributesList()) {
              System.out.printf("\n\t\t\tName: %s\n", attribute.getName());
              System.out.printf("\t\t\tConfidence: %s\n", attribute.getConfidence());
              System.out.printf("\t\t\tValue: %s\n", attribute.getValue());
            }
          }

          // Optional. Attributes in the track level.
          for (DetectedAttribute trackAttribute : track.getAttributesList()) {
            System.out.printf("\n\t\tName : %s\n", trackAttribute.getName());
            System.out.printf("\t\tConfidence : %s\n", trackAttribute.getConfidence());
            System.out.printf("\t\tValue : %s\n", trackAttribute.getValue());
          }
        }

        // All video segments where the recognized logo appears. There might be multiple instances
        // of the same logo class appearing in one VideoSegment.
        for (VideoSegment segment : logoRecognitionAnnotation.getSegmentsList()) {
          System.out.printf(
              "\n\tStart Time Offset : %s.%s\n",
              segment.getStartTimeOffset().getSeconds(), segment.getStartTimeOffset().getNanos());
          System.out.printf(
              "\tEnd Time Offset : %s.%s\n",
              segment.getEndTimeOffset().getSeconds(), segment.getEndTimeOffset().getNanos());
        }
      }
    }
  }
}

Node.js

Per autenticarti a Video Intelligence, configura le Credenziali predefinite dell'applicazione. Per ulteriori informazioni, consulta Configurare l'autenticazione per un ambiente di sviluppo locale. 

/**
 * TODO(developer): Uncomment these variables before running the sample.
 */
// const localFilePath = 'path/to/your/video.mp4'

// Imports the Google Cloud client libraries
const Video = require('@google-cloud/video-intelligence');
const fs = require('fs');

// Instantiates a client
const client = new Video.VideoIntelligenceServiceClient();

// Performs asynchronous video annotation for logo recognition on a file.
async function detectLogo() {
  const inputContent = fs.readFileSync(localFilePath).toString('base64');

  // Build the request with the input content and logo recognition feature.
  const request = {
    inputContent: inputContent,
    features: ['LOGO_RECOGNITION'],
  };

  // Make the asynchronous request
  const [operation] = await client.annotateVideo(request);

  // Wait for the results
  const [response] = await operation.promise();

  // Get the first response, since we sent only one video.
  const annotationResult = response.annotationResults[0];
  for (const logoRecognitionAnnotation of annotationResult.logoRecognitionAnnotations) {
    const entity = logoRecognitionAnnotation.entity;
    // Opaque entity ID. Some IDs may be available in
    // [Google Knowledge Graph Search API](https://developers.google.com/knowledge-graph/).
    console.log(`Entity Id: ${entity.entityId}`);
    console.log(`Description: ${entity.description}`);

    // All logo tracks where the recognized logo appears.
    // Each track corresponds to one logo instance appearing in consecutive frames.
    for (const track of logoRecognitionAnnotation.tracks) {
      console.log(
        `\n\tStart Time Offset: ${track.segment.startTimeOffset.seconds}.${track.segment.startTimeOffset.nanos}`
      );
      console.log(
        `\tEnd Time Offset: ${track.segment.endTimeOffset.seconds}.${track.segment.endTimeOffset.nanos}`
      );
      console.log(`\tConfidence: ${track.confidence}`);

      // The object with timestamp and attributes per frame in the track.
      for (const timestampedObject of track.timestampedObjects) {
        // Normalized Bounding box in a frame, where the object is located.
        const normalizedBoundingBox = timestampedObject.normalizedBoundingBox;
        console.log(`\n\t\tLeft: ${normalizedBoundingBox.left}`);
        console.log(`\t\tTop: ${normalizedBoundingBox.top}`);
        console.log(`\t\tRight: ${normalizedBoundingBox.right}`);
        console.log(`\t\tBottom: ${normalizedBoundingBox.bottom}`);
        // Optional. The attributes of the object in the bounding box.
        for (const attribute of timestampedObject.attributes) {
          console.log(`\n\t\t\tName: ${attribute.name}`);
          console.log(`\t\t\tConfidence: ${attribute.confidence}`);
          console.log(`\t\t\tValue: ${attribute.value}`);
        }
      }

      // Optional. Attributes in the track level.
      for (const trackAttribute of track.attributes) {
        console.log(`\n\t\tName: ${trackAttribute.name}`);
        console.log(`\t\tConfidence: ${trackAttribute.confidence}`);
        console.log(`\t\tValue: ${trackAttribute.value}`);
      }
    }

    // All video segments where the recognized logo appears.
    // There might be multiple instances of the same logo class appearing in one VideoSegment.
    for (const segment of logoRecognitionAnnotation.segments) {
      console.log(
        `\n\tStart Time Offset: ${segment.startTimeOffset.seconds}.${segment.startTimeOffset.nanos}`
      );
      console.log(
        `\tEnd Time Offset: ${segment.endTimeOffset.seconds}.${segment.endTimeOffset.nanos}`
      );
    }
  }
}

detectLogo();

Python

Per autenticarti a Video Intelligence, configura le Credenziali predefinite dell'applicazione. Per ulteriori informazioni, consulta Configurare l'autenticazione per un ambiente di sviluppo locale. 

import io

from google.cloud import videointelligence


def detect_logo(local_file_path="path/to/your/video.mp4"):
    """Performs asynchronous video annotation for logo recognition on a local file."""

    client = videointelligence.VideoIntelligenceServiceClient()

    with io.open(local_file_path, "rb") as f:
        input_content = f.read()
    features = [videointelligence.Feature.LOGO_RECOGNITION]

    operation = client.annotate_video(
        request={"features": features, "input_content": input_content}
    )

    print("Waiting for operation to complete...")
    response = operation.result()

    # Get the first response, since we sent only one video.
    annotation_result = response.annotation_results[0]

    # Annotations for list of logos detected, tracked and recognized in video.
    for logo_recognition_annotation in annotation_result.logo_recognition_annotations:
        entity = logo_recognition_annotation.entity

        # Opaque entity ID. Some IDs may be available in [Google Knowledge Graph
        # Search API](https://developers.google.com/knowledge-graph/).
        print("Entity Id : {}".format(entity.entity_id))

        print("Description : {}".format(entity.description))

        # All logo tracks where the recognized logo appears. Each track corresponds
        # to one logo instance appearing in consecutive frames.
        for track in logo_recognition_annotation.tracks:
            # Video segment of a track.
            print(
                "\n\tStart Time Offset : {}.{}".format(
                    track.segment.start_time_offset.seconds,
                    track.segment.start_time_offset.microseconds * 1000,
                )
            )
            print(
                "\tEnd Time Offset : {}.{}".format(
                    track.segment.end_time_offset.seconds,
                    track.segment.end_time_offset.microseconds * 1000,
                )
            )
            print("\tConfidence : {}".format(track.confidence))

            # The object with timestamp and attributes per frame in the track.
            for timestamped_object in track.timestamped_objects:
                # Normalized Bounding box in a frame, where the object is located.
                normalized_bounding_box = timestamped_object.normalized_bounding_box
                print("\n\t\tLeft : {}".format(normalized_bounding_box.left))
                print("\t\tTop : {}".format(normalized_bounding_box.top))
                print("\t\tRight : {}".format(normalized_bounding_box.right))
                print("\t\tBottom : {}".format(normalized_bounding_box.bottom))

                # Optional. The attributes of the object in the bounding box.
                for attribute in timestamped_object.attributes:
                    print("\n\t\t\tName : {}".format(attribute.name))
                    print("\t\t\tConfidence : {}".format(attribute.confidence))
                    print("\t\t\tValue : {}".format(attribute.value))

            # Optional. Attributes in the track level.
            for track_attribute in track.attributes:
                print("\n\t\tName : {}".format(track_attribute.name))
                print("\t\tConfidence : {}".format(track_attribute.confidence))
                print("\t\tValue : {}".format(track_attribute.value))

        # All video segments where the recognized logo appears. There might be
        # multiple instances of the same logo class appearing in one VideoSegment.
        for segment in logo_recognition_annotation.segments:
            print(
                "\n\tStart Time Offset : {}.{}".format(
                    segment.start_time_offset.seconds,
                    segment.start_time_offset.microseconds * 1000,
                )
            )
            print(
                "\tEnd Time Offset : {}.{}".format(
                    segment.end_time_offset.seconds,
                    segment.end_time_offset.microseconds * 1000,
                )
            )

Linguaggi aggiuntivi

C#: segui le istruzioni di configurazione per C# nella pagina delle librerie client e poi consulta la documentazione di riferimento di Video Intelligence per .NET.

PHP: segui le istruzioni di configurazione di PHP riportate nella pagina delle librerie client e consulta la documentazione di riferimento di Video Intelligence per PHP.

Ruby: segui le istruzioni di configurazione di Ruby riportate nella pagina delle librerie client e consulta la documentazione di riferimento di Video Intelligence per Ruby.