This is an old revision of the document!
Table of Contents
Image Classification Game
This Snap! game uses Nvidia Jetson capability to classify images.
Offline Snap! downloading
Please download and open Offline version of Snap! for our project. Go to https://snap.berkeley.edu/offline and follow the steps.
Snap! files' downloading
Please open the link Classification Game to download our project on your computer. Probably you would see the xml in raw format. Click the right button of your mouse and save it on the disk.
Web camera Image in Snap!
You can get picture from your web camera in Snap!.
Connection to Jetson from Snap!
If you have not imported it yet, please download jetson blocks and import it to your Snap! project.
Enable Javascript Extensions for following blocks.
You need ip address of Nvidia Jetson. You can use ifconfig command in a terminal to get ip address.
Response from classification
Here we will send video snap on stage to Jetson for processing. Jetson will respond back class name, confidence value and class ID.
Only class name and confidence value will be used in this example. This project does not use class ID .
- Use get response from Jetson block to send image , and get class name and confidence value.
- First input slot is for jetson variable that stores websocket data.
- Second input slot is for costume you want to be classified by Nvidia Jetson.
Class name and confidence value
This section will demonstrate how to handle response variable to access class name and confidence value.
You can create custom blocks, to get class name get class name from response and to get confidence get confidence from response .
Speech functionality
Speech functionality is available as a library in Snap!. Select export libraries from settings then choose speech module .
Repeat block for game
Last step is adding loop for the game.
This example used repeat until block to break loop when space key pressed.
You can download full game from Github page of EOLab-HSRW.