To train an object detection model from scratch will require long hours of model training. To save time, the simplest approach would be to use an already trained model and retrain it to detect your custom objects. This process is called “transfer learning”.
Object detection model performance continues to improve. As of 2020, MobilenetV2 is the fastest object detection model which can run in a web browser.
There are other models for object detection such as YOLO and more. You can find a detailed object detection model review here.
MobilenetV2 is a model based on TensorFlow, therefore we will execute…
How to build your object detection web-app with retrained MobilenetV2 model in 30 minutes.
Creating web apps for object detection is easy and fun. It can be done with frameworks like pl5 which are based on ported models trained on coco data sets (coco-ssd), and running the TensorFlow.js engine under the hood.
It becomes a bit complicated if you try to train your model yourself on custom data sets. For example: Creating an app which detects your favorite toy.
In this article I want to show you how to create TURN server and use it in peer.js framework.
TURN server is found to be most useful in creating peer to peer video chat, where it is used to relay data traffic between the two peers for frameworks based on WebRTC . More about WebRTC you can find it here.
Commonly used free STUN servers will try to resolve IP public addresses of chat participants .Unfortunately, in most cases, where one of the peers is behind NAT, STUN server fail and we left with only one option, and it is…
Researcher in #Real time #VR/#AR classification #AI navigation and #Object recognition and detection.