Multimodal Embeddings API

import vertexai
from vertexai.vision_models import Image, MultiModalEmbeddingModel

# TODO(developer): Update values for project_id, image_path & contextual_text
vertexai.init(project=project_id, location="us-central1")

model = MultiModalEmbeddingModel.from_pretrained("multimodalembedding")
image = Image.load_from_file(image_path)

embeddings = model.get_embeddings(
    image=image,
    contextual_text=contextual_text,
)
print(f"Image Embedding: {embeddings.image_embedding}")
print(f"Text Embedding: {embeddings.text_embedding}")

/**
 * TODO(developer): Uncomment these variables before running the sample.\
 * (Not necessary if passing values as arguments)
 */
// const project = 'YOUR_PROJECT_ID';
// const location = 'YOUR_PROJECT_LOCATION';
// const bastImagePath = "YOUR_BASED_IMAGE_PATH"
// const textPrompt = 'YOUR_TEXT_PROMPT';
const aiplatform = require('@google-cloud/aiplatform');

// Imports the Google Cloud Prediction service client
const {PredictionServiceClient} = aiplatform.v1;

// Import the helper module for converting arbitrary protobuf.Value objects.
const {helpers} = aiplatform;

// Specifies the location of the api endpoint
const clientOptions = {
  apiEndpoint: 'us-central1-aiplatform.googleapis.com',
};
const publisher = 'google';
const model = 'multimodalembedding@001';

// Instantiates a client
const predictionServiceClient = new PredictionServiceClient(clientOptions);

async function predictImageFromImageAndText() {
  // Configure the parent resource
  const endpoint = `projects/${project}/locations/${location}/publishers/${publisher}/models/${model}`;

  const fs = require('fs');
  const imageFile = fs.readFileSync(baseImagePath);

  // Convert the image data to a Buffer and base64 encode it.
  const encodedImage = Buffer.from(imageFile).toString('base64');

  const prompt = {
    text: textPrompt,
    image: {
      bytesBase64Encoded: encodedImage,
    },
  };
  const instanceValue = helpers.toValue(prompt);
  const instances = [instanceValue];

  const parameter = {
    sampleCount: 1,
  };
  const parameters = helpers.toValue(parameter);

  const request = {
    endpoint,
    instances,
    parameters,
  };

  // Predict request
  const [response] = await predictionServiceClient.predict(request);
  console.log('Get image embedding response');
  const predictions = response.predictions;
  console.log('\tPredictions :');
  for (const prediction of predictions) {
    console.log(`\t\tPrediction : ${JSON.stringify(prediction)}`);
  }
}

await predictImageFromImageAndText();

import com.google.cloud.aiplatform.v1beta1.EndpointName;
import com.google.cloud.aiplatform.v1beta1.PredictResponse;
import com.google.cloud.aiplatform.v1beta1.PredictionServiceClient;
import com.google.cloud.aiplatform.v1beta1.PredictionServiceSettings;
import com.google.gson.Gson;
import com.google.gson.JsonObject;
import com.google.protobuf.InvalidProtocolBufferException;
import com.google.protobuf.Value;
import com.google.protobuf.util.JsonFormat;
import java.io.IOException;
import java.nio.charset.StandardCharsets;
import java.nio.file.Files;
import java.nio.file.Paths;
import java.util.ArrayList;
import java.util.Base64;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

public class PredictImageFromImageAndTextSample {

  public static void main(String[] args) throws IOException {
    // TODO(developer): Replace this variable before running the sample.
    String project = "YOUR_PROJECT_ID";
    String textPrompt = "YOUR_TEXT_PROMPT";
    String baseImagePath = "YOUR_BASE_IMAGE_PATH";

    // Learn how to use text prompts to update an image:
    // http://cloud.go888ogle.com.fqhub.com/vertex-ai/docs/generative-ai/image/edit-images
    Map<String, Object> parameters = new HashMap<String, Object>();
    parameters.put("sampleCount", 1);

    String location = "us-central1";
    String publisher = "google";
    String model = "multimodalembedding@001";

    predictImageFromImageAndText(
        project, location, publisher, model, textPrompt, baseImagePath, parameters);
  }

  // Update images using text prompts
  public static void predictImageFromImageAndText(
      String project,
      String location,
      String publisher,
      String model,
      String textPrompt,
      String baseImagePath,
      Map<String, Object> parameters)
      throws IOException {
    final String endpoint = String.format("%s-aiplatform.googleapis.com:443", location);
    final PredictionServiceSettings predictionServiceSettings =
        PredictionServiceSettings.newBuilder().setEndpoint(endpoint).build();

    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests.
    try (PredictionServiceClient predictionServiceClient =
        PredictionServiceClient.create(predictionServiceSettings)) {
      final EndpointName endpointName =
          EndpointName.ofProjectLocationPublisherModelName(project, location, publisher, model);

      // Convert the image to Base64
      byte[] imageData = Base64.getEncoder().encode(Files.readAllBytes(Paths.get(baseImagePath)));
      String encodedImage = new String(imageData, StandardCharsets.UTF_8);

      JsonObject jsonInstance = new JsonObject();
      jsonInstance.addProperty("text", textPrompt);
      JsonObject jsonImage = new JsonObject();
      jsonImage.addProperty("bytesBase64Encoded", encodedImage);
      jsonInstance.add("image", jsonImage);

      Value instanceValue = stringToValue(jsonInstance.toString());
      List<Value> instances = new ArrayList<>();
      instances.add(instanceValue);

      Gson gson = new Gson();
      String gsonString = gson.toJson(parameters);
      Value parameterValue = stringToValue(gsonString);

      PredictResponse predictResponse =
          predictionServiceClient.predict(endpointName, instances, parameterValue);
      System.out.println("Predict Response");
      System.out.println(predictResponse);
      for (Value prediction : predictResponse.getPredictionsList()) {
        System.out.format("\tPrediction: %s\n", prediction);
      }
    }
  }

  // Convert a Json string to a protobuf.Value
  static Value stringToValue(String value) throws InvalidProtocolBufferException {
    Value.Builder builder = Value.newBuilder();
    JsonFormat.parser().merge(value, builder);
    return builder.build();
  }
}

import vertexai

from vertexai.vision_models import MultiModalEmbeddingModel, Video

# TODO(developer): Update values for project_id,
#               video_path, contextual_text, dimension, video_segment_config
vertexai.init(project=project_id, location="us-central1")

model = MultiModalEmbeddingModel.from_pretrained("multimodalembedding")
video = Video.load_from_file(video_path)

embeddings = model.get_embeddings(
    video=video,
    video_segment_config=video_segment_config,
    contextual_text=contextual_text,
    dimension=dimension,
)

# Video Embeddings are segmented based on the video_segment_config.
print("Video Embeddings:")
for video_embedding in embeddings.video_embeddings:
    print(
        f"Video Segment: {video_embedding.start_offset_sec} - {video_embedding.end_offset_sec}"
    )
    print(f"Embedding: {video_embedding.embedding}")

print(f"Text Embedding: {embeddings.text_embedding}")

import vertexai

from vertexai.vision_models import Image, MultiModalEmbeddingModel, Video

# TODO(developer): Update values for project_id,
#            image_path, video_path, contextual_text, video_segment_config
vertexai.init(project=project_id, location="us-central1")

model = MultiModalEmbeddingModel.from_pretrained("multimodalembedding")
image = Image.load_from_file(image_path)
video = Video.load_from_file(video_path)

embeddings = model.get_embeddings(
    image=image,
    video=video,
    video_segment_config=video_segment_config,
    contextual_text=contextual_text,
    dimension=dimension,
)

print(f"Image Embedding: {embeddings.image_embedding}")

# Video Embeddings are segmented based on the video_segment_config.
print("Video Embeddings:")
for video_embedding in embeddings.video_embeddings:
    print(
        f"Video Segment: {video_embedding.start_offset_sec} - {video_embedding.end_offset_sec}"
    )
    print(f"Embedding: {video_embedding.embedding}")

print(f"Text Embedding: {embeddings.text_embedding}")