Hack:Pack

Hack:Pack is an AI mobile scanner to create rapid prototypes of object-triggered animations for A/B testing with target populations: it automates the process of 2D/3D object recognition using leading edge AI technology named "deep learning".".
Hack:Pack's Convolutional Neural Network (CNN) can be trained to recognize new objects ("datasets") from a video directly recorded from the mobile app, and animations can also be directly overlayed from the app's library.
The Hack:Pack app comes with a mirror web platform on which users can further refine their animation scenario to optimize user experience.
Main features include:
* Create a training dataset from a 1min video of any 2D/3D object recorded directly with the app
* Add your new dataset to the library of recognized objects, in public or private mode
* Select animations to be triggered by your trigger object from URL's, videos (MP4), 2D pictures (GIF, PNG, JPG) or 3D models (FBX, GLB)
* Access all functions of the app without login or password
* Make your trigger datasets private by creating a login
* Connect to the mirror web Hack:Pack platform to refine your overlays, improve user experience and access the statistics of scanning of your triggers.

更新履歴
- Minor Bug Fix.
- UI Enhancement.
- Performance Improvement.