Project Info
This project did not submit a demo video on Devpost.
What it does
What our project does is introduce ASL letters, words, and numbers to you in a flashcard manner.
How we built it
We built our project with React, Vite, and TensorFlowJS.
Challenges we ran into
Some challenges we ran into included issues with git commits and merging. Over the course of our project we made mistakes while resolving merge conflicts which resulted in a large part of our project being almost discarded. Luckily we were able to git revert back to the correct version but time was misused regardless. With our TensorFlow model we had trouble reading the input/output and getting webcam working.
Accomplishments we're proud of
We are proud of the work we got done in the time frame of this hackathon with our skill level. Out of the workshops we attended I think we learned a lot and can't wait to implement them in future projects!
What we learned
Over the course of this hackathon we learned that it is important to clearly define our project scope ahead of time. We spent a lot of our time on day 1 thinking about what we could do with the sponsor technologies and should have looked into them more in depth before the hackathon.
What's next
We would like to train our own ASL image detection model so that people can practice at home in real time. Additionally we would like to transcribe their signs into plaintext and voice so that they can confirm what they are signing. Expanding our project scope beyond ASL to other languages is also something we wish to do.
vision-talks
CalHacks 11.0 project
Card image sources:
https://www.printablee.com/postpic/2013/07/basic-baby-sign-language-words-chart_392258.jpg
https://www.startasl.com/wp-content/uploads/sign-language-flashcards.pdf
https://www.slideshare.net/slideshow/numbers-in-asl/70722207#29
Analysis
View
Metric
- 33
- 19
- 2
Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.
Technology
- CSSIn code
- HTMLIn code
- JavaScriptIn code
- ReactIn code
- TypeScriptIn code
5 of 5 appear in the indexed code.
AI coding agents
No AI coding agent signals were found in this repository.
Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.
Codebase size
Source size
31 KB
Source files
32
Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.
Repository
aghosh376/vision-talks
97 files · 2.9 MB · @ 32adb9f
Structure
Interface
15 files · 15%Screens, components and styles rendered to the user.
Application logic
11 files · 11%Domain rules, services and shared utilities.
Supporting
Layers are inferred from where files sit in the tree, not from reading the code. A project that names its directories unconventionally will read oddly here — open the file browser to check anything the diagram implies.
Languages
- TypeScript42%
- CSS33%
- JavaScript13%
- Markdown8%
- YAML3%
- HTML2%
Share of indexed source by file size. Binary and vendored files are excluded.
Dependencies
vision/package.json
npm · 19- @tensorflow-models/coco-ssd
- @tensorflow-models/hand-pose-detection
- @tensorflow-models/handpose
- @tensorflow/tfjs
- @tensorflow/tfjs-backend-webgl
- react
- react-dom
- react-router-dom
- +11 more
vision-talks/package.json
npm · 3- react-router-dom
- +2 more
Declared in the repository’s manifests at the indexed commit. A declared package is not proof it is used, and runtime dependencies are listed first.
Feature verification
ASL letter flashcardsVerified
Introduces ASL letters in a flashcard manner
Claimed on Devposthigh confidencevision/src/pages/DictPage.tsx:5— cardData array maps every letter A-Z to an image and renders a Card component per lettervision/src/components/Card.tsx:9— Card component implements a flip-on-click flashcard UI showing letter then imagevision/public/assets/pictures/A.png— corresponding image assets exist for the referenced letters
ASL number flashcardsVerified
Introduces ASL numbers in a flashcard manner
Claimed on Devposthigh confidencevision/src/pages/DictPage.tsx:32— cardData includes number entries 0-9, 10, 100, 1000 reusing the same Card componentvision/public/assets/pictures/1000.png— corresponding number image assets exist
ASL word flashcardsVerified
Introduces ASL words in a flashcard manner
Claimed on Devposthigh confidencevision/src/pages/WordPage.tsx:5— wordCardData array with words like Eat, Hello, Goodbye, Please rendered via Card components
Built with ReactVerified
Built with React
Claimed on Devposthigh confidencevision/package.json:18— react and react-dom listed as dependencies, and .tsx components use React throughout
Built with TensorFlowJSVerified
Built with TensorFlowJS
Claimed on Devposthigh confidencevision/package.json:16— @tensorflow/tfjs and related model packages listed as dependenciesvision/src/components/ObjectDetection.tsx:1— imports handpose, hand-pose-detection, tfjs-backend-webgl, and coco-ssd and uses them for webcam-based detection
Built with ViteVerified
Built with Vite
Claimed on Devposthigh confidencevision/vite.config.ts— vite config file present and vite is a devDependency/scripts use vite
Real-time ASL sign detection/practice via webcamCode-supported
Train own ASL image detection model so people can practice at home in real time and transcribe signs into plaintext/voice (What's next section, implies partial webcam detection exists)
Claimed on Devpostmedium confidencevision/src/components/ObjectDetection.tsx:47— webcam capture and detection loop exist, but predictObject uses cocoSsd (generic object detection), not an ASL-specific hand sign classifier; no ASL letter/word recognition or plaintext/voice transcription logic foundvision/src/pages/Detect.tsx:1— Detect page wires ObjectDetection into a route labeled 'Hand Pose Detection', though handpose/hand-pose-detection imports in ObjectDetection.tsx are unused in the component logic
Custom-trained ASL detection modelClaimed only
Train our own ASL image detection model
Claimed on Devposthigh confidenceExpand beyond ASL to other languagesClaimed only
Expanding project scope beyond ASL to other languages
Claimed on Devposthigh confidenceTranscribe signs into plaintext and voiceClaimed only
Transcribe their signs into plaintext and voice so users can confirm what they are signing
Claimed on Devposthigh confidence
An AI agent derived these features from the project’s Devpost page and readme, then searched the code for each one. Verified features are backed by cited code; claimed-only features had no supporting code, which is not by itself proof a feature is missing.
Export this project's context (description, README, evidence, key source files) to chat with an AI agent elsewhere.