tfjs-models VS samples

I am wondering, will this also work with tensoflow.js or only tflite? I'd like to use this hand pose estimation from mediapipe: https://github.com/tensorflow/tfjs-models/tree/master/hand-pose-detection

I got the basic concept working using TensorFlow QnA model but the answers don't seem very accurate. Infact, unless you specifically ask the exact question, you dont get the right answer. Its not intelligent enough because the entire PDF content becomes a bag of words instead of having context to those words. for eg. When someone types "languages", it should search within a section named Languages.

I think you are beginning to connect the dots by now 😉 What we will do is to build a small proof-of-concept (POC) by writing a simple react app and hook up a pre-trained tensorflow.js model, The text toxicity model to "moderate" the user's text input and show a notification of what's wrong with it, a text toxicity meter if you will...

TensorFlow.js is a JavaScript library developers can use to run pre-trained machine-learning models in the browser. The library has a variety of models for tasks such as object identification and language processing. One of these models is the text toxicity detection model.

TensorFlow has a toxicity classification model available. I don't find it to work that well, but it does something. Not sure if it works in other languages than English. Tried the demo with some offensive Dutch and it passed it.

No clue if it will work on drawn images. It depends on what the model was trained on. The models github is a good place to get started on figuring that out: https://github.com/tensorflow/tfjs-models/tree/body-pix-v2.0.4/body-pix

You probably want to check out something like tensorflow https://github.com/tensorflow/tfjs-models where you can build and test your models.

I decided to start with the toxicity model OOTB demo code for simplicity. You can find this code in their Github repo.

I record video from a webcam using MediaRecorder 100ms segments in the way described here. When the piece is ready, I send it using socket.io to the server (on a node.js 6.3.0 server). The server writes this piece to the database and gives it to everyone else on the socket. It is known for sure that the correct data is coming to the server, I can save any piece and start the video (then I glue them together and get the recording). The problem is that when the data is returned back to the client, it is invalid.

Thank you so far :) the "complex data processing" is a webpacked DSP hardware simulator, which has to exist in the global scope as it manipulates the DOM and cannot re-instantiate within the AudioWorkletGlobalScope. I have found this, which uses MediaStreamTrackProcessor, MediaStreamTrackGenerator and TransformStream to expose the microphone samples to the global scope and then "cheats" by sending them to an audio HTML component instead of directly to the audiocontext destination, mimicing actual realtime processing. This is what I was looking for, however:

WebRTC: https://webrtc.github.io/samples/

Demos are often snapshots in time, created to showcase an aspect of technology. They are typically small, self-contained, and single-purpose. Demonstrative projects can range from the practical stress-testing of a particular feature to creative, wacky ways to use them. Examples include WebRTC samples and the Dependabot Demo.

This repo includes all the tutorials that are on https://webrtc.github.io/samples/. The aim of the project is to convert all the sample code to a Svelte app.

Maybe you can identify which part of WebRTC is failing using some test webpages like https://webrtc.github.io/samples/

Here's a few samples. The video chat example will be very useful to you.