Project
# | Title | Team Members | TA | Documents | Sponsor |
---|---|---|---|---|---|
27 | Real-Time Sign Language Translator |
Gene Lee Kaelan To |
Bonhyun Ku | design_document1.pdf design_document2.pdf final_paper1.pdf presentation1.pptx proposal1.pdf |
|
# Real-Time Sign Language Translator Team Members: - Gene Lee (genel2) - Kaelan To (kto3) # Problem (Describe the problem you want to solve and motivate the need.) Technology is improving rapidly, and with that, it serves the purpose of making our lives easier. We want to leverage the technology available to us and further integrate those with disabilities more into our society; specifically deaf individuals in an academic setting. When we imagine students with hearing impairments working with others, we think of the other students having to wait for the student to type out their thoughts. However, brainstorming/bouncing off ideas requires rapid discussion in order to spark good ideas. Sometimes typing may not be as fast (which might hinder the group) or even may not be accessible to students (especially in K-12) in classroom settings. # Solution (Describe your design at a high-level, how it solves the problem, and introduce the subsystems of your project.) We propose a portable real-time sign language translator to solve this problem. We would utilize computer vision to differentiate the hand signs and feed the visual input to a microcontroller and give audio feedback (sound translation of the hand sign). This portable system would assist them in communicating with teachers, but most definitely help deaf students work in a team with other students efficiently. # Solution Components Explain what the subsystem does. Explicitly list what sensors/components you will use in this subsystem. Include part numbers. ## Visual Input A camera will be used to read input from the user and send that information to the central processor. ## Central Processor The central processor will decode the input from the camera and send that information to the audio component. We plan on using a pose estimation library on a Raspberry Pi to process the input. ## Audio Speakers will take the decoded output from the central processor and play it out from the encasing. ## Power The battery should have a large enough charge to last about 8 hours, which is a typical school day from grades K-12. They also need to be small enough to fit in a relatively small encasing to be more portable. ## Encasing All the components are expected to fit inside a casing that would be portable enough for a student to carry around from classroom to classroom without much trouble. The goal is to be able to fit all components inside a 15x15x15cm cube or other encasing of equal or smaller volume. # Criterion For Success (Describe high-level goals that your project needs to achieve to be effective. These goals need to be clearly testable and not subjective.) Able to identify sign language and translate into English in real-time (threshold set to be within 0.5 seconds) Able to identify signing at a moderate/conversational level speed (threshold to be set after more discussion/research) System is lightweight/portable (not hard to carry around) Battery lifetime of at least one school day(8 hours) |