Beta Presentation Image Recognition, Annotation and Validation Mobile Application The Capstone Experience Team Whirlpool Shruti Avutapalli Jessica Clappison Jackie Li Savanna Pinkoski Jack Turak Department of Computer Science and Engineering Michigan State University From Students… Fall 2018 …to Professionals
Project Overview Problem: • Lack of realistic training photos to achieve high object recognition accuracy. Project Objective: • Improve Yummly’s image recognition data model. • Collect annotated photos from Whirlpool employees. • Evaluate correctness of annotations through validation interface. The Capstone Experience Team Whirlpool Beta Presentation 2
System Architecture User Creation/Sign-In Authentication Image Storage TensorFlow Lite Android Storage Firebase Backend API Admin Dashboard Cloud iOS Functions The Capstone Experience Team Whirlpool Beta Presentation 6
Image Capture Interface The Capstone Experience Team Whirlpool Beta Presentation 7
Annotation Interface The Capstone Experience Team Whirlpool Beta Presentation 8
Annotation Interface: Completed The Capstone Experience Team Whirlpool Beta Presentation 9
Validation Interface The Capstone Experience Team Whirlpool Beta Presentation 10
Admin Dashboard The Capstone Experience Team Whirlpool Beta Presentation 10
Admin Dashboard The Capstone Experience Team Whirlpool Beta Presentation 10
What’s left to do? • Testing User feedback Bug fixes Edge cases • UI Design Finalization • Documentation The Capstone Experience Team Whirlpool Beta Presentation 10
Questions? ? ? ? ? ? ? ? ? ? The Capstone Experience Team Whirlpool Beta Presentation 11
Recommend
More recommend