Webcam as Greengrass component?

Hi, there are multiple ways to achieve what you are looking for. The one I would advise for is the following:

1. Deploy AWS IoT Greengrass core on Computer A
2. Deploy a component on Computer A that accesses the camera and perform object detection on the frames captured from the camera. You can capture video or frames via OpenCV. The user on Computer A will likely need to allow the application to use the webcam

import cv2
import numpy as np
cap = cv2.VideoCapture(0)
while True:
    ret, frame = cap.read()
    # 1. perform object detection on frame
    # 2. annotate the image and write it to file
    # 3. upload the file to S3
cap.release()
cv2.destroyAllWindows()

3. Upload the frames with the object detection annotations to an S3 bucket, eg using the StreamManager component or uploading directly to S3 via an AWS SDK
4. Write an authenticated API to access the annotated frames from the S3 bucket
5. Write a react application hosted on S3+CloudFront (you can use AWS Amplify) [1]. The application access the API to retrieve the annotated frames. You might also consider adding user authentication to restrict who has access to the application
6. From computer B, a user opens the browser to the application URL, authenticates, and access the image frames

If the analysis of the frames is performed only when user on computer B performs an action, you might need to add a command and control feature to your solution, for example using AWS IoT.

Useful links: [1] https://aws.amazon.com/blogs/mobile/build-a-photo-gallery-react-app-using-amplify-studios-new-file-storage-capabilities/

It is possible to use AWS IoT Greengrass to deploy a real-time object detection application that uses the webcam of computer A and displays the results on the monitor of computer B. Here's a high-level overview of how you can achieve this:

1. Connect the webcam of computer A to the internet and make sure it is accessible via a network URL or IP address.

2. Create an AWS IoT Greengrass group and a core device.

3. Create a Greengrass Lambda function that uses the webcam of computer A as its video source. The function should use a library such as OpenCV or TensorFlow to perform real-time object detection on the video stream.

4. Create another Lambda function that receives the object detection results from the first function and displays them on the monitor of computer B. This function can use a library such as PyQt or OpenCV to display the results on the monitor.

5. Deploy the Greengrass group and the Lambda functions to the core device.

6. Configure the Lambda functions to communicate with each other using the Greengrass messaging API.

7. Test the application by running the object detection Lambda function and checking that the results are displayed on the monitor of computer B.