# arjunraj-01/sign-language architecture

> Architecture diagram of the GitHub repository arjunraj-01/sign-language, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-07). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/arjunraj-01/sign-language
- Repository: https://github.com/arjunraj-01/sign-language

## Overview

SignBridge is a webcam-based ASL alphabet recognizer that turns detected gestures into text. Its main user workflow runs from camera frames through hand detection, image classification and landmark-based disambiguation, then stabilizes predictions and lets users build words with controls. The repository also includes a separate dataset preparation and TensorFlow training/evaluation path. The sampled inference code shows detector and classifier behavior; some app smoothing details and several documented training capabilities are not fully shown, so uncertain links are omitted or marked unverified.

Main entry points: [`app.py`](https://github.com/arjunraj-01/sign-language/blob/main/app.py) (Streamlit app), [`train.py`](https://github.com/arjunraj-01/sign-language/blob/main/train.py) (Model training), [`src/inference.py`](https://github.com/arjunraj-01/sign-language/blob/main/src/inference.py) (Inference engine), [`src/preprocessing.py`](https://github.com/arjunraj-01/sign-language/blob/main/src/preprocessing.py) (Dataset loading)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_interaction["User interaction"]
  node_app["Streamlit app<br/>[app.py]"]
  node_words["Word builder<br/>[word_builder.py]"]
  node_smoother["Prediction smoother<br/>[smoothing.py]"]
end

subgraph group_recognition["Live recognition"]
  node_inference["Inference engine<br/>[inference.py]"]
  node_detector["Hand detector<br/>[hand_detection.py]"]
  node_rules["Landmark disambiguation<br/>[landmark_rules.py]"]
end

subgraph group_learning["Model preparation"]
  node_dataset["Dataset loading<br/>[preprocessing.py]"]
  node_model["CNN model builders<br/>[model.py]"]
  node_training["Model training<br/>[train.py]"]
  node_evaluation["Model evaluation<br/>[evaluate.py]"]
end

subgraph group_support["Shared settings"]
  node_config["Runtime configuration<br/>[config.py]"]
end

node_user(("Signer"))
node_webcam["Webcam"]
node_data[("ASL image dataset")]
node_trainedmodel[("Trained classifier")]

node_user -->|"operates"| node_app
node_webcam -->|"provides frames"| node_app
node_app -->|"requests predictions"| node_inference
node_inference -->|"detects hand"| node_detector
node_inference -->|"disambiguates"| node_rules
node_inference -.->|"feeds predictions"| node_smoother
node_app -->|"updates text"| node_words
node_app -.->|"stabilizes output"| node_smoother
node_data -->|"loads images"| node_dataset
node_training -->|"loads splits"| node_dataset
node_training -->|"builds classifier"| node_model
node_training -->|"saves weights"| node_trainedmodel
node_evaluation -->|"loads data"| node_dataset
node_evaluation -.->|"evaluates"| node_trainedmodel
node_training -->|"reads settings"| node_config
node_dataset -->|"reads settings"| node_config
node_app -.->|"uses defaults"| node_config

click node_app "https://github.com/arjunraj-01/sign-language/blob/main/app.py"
click node_words "https://github.com/arjunraj-01/sign-language/blob/main/src/word_builder.py"
click node_smoother "https://github.com/arjunraj-01/sign-language/blob/main/src/smoothing.py"
click node_inference "https://github.com/arjunraj-01/sign-language/blob/main/src/inference.py"
click node_detector "https://github.com/arjunraj-01/sign-language/blob/main/src/hand_detection.py"
click node_rules "https://github.com/arjunraj-01/sign-language/blob/main/src/landmark_rules.py"
click node_dataset "https://github.com/arjunraj-01/sign-language/blob/main/src/preprocessing.py"
click node_model "https://github.com/arjunraj-01/sign-language/blob/main/src/model.py"
click node_training "https://github.com/arjunraj-01/sign-language/blob/main/train.py"
click node_evaluation "https://github.com/arjunraj-01/sign-language/blob/main/evaluate.py"
click node_config "https://github.com/arjunraj-01/sign-language/blob/main/src/config.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_app,node_words,node_smoother toneBlue
class node_inference,node_detector,node_rules,node_data,node_trainedmodel toneAmber
class node_dataset,node_model,node_training,node_evaluation toneMint
class node_config toneRose
class node_user,node_webcam toneIndigo
```

## Components

### User interaction

- [Streamlit app](https://github.com/arjunraj-01/sign-language/blob/main/app.py): `app.py`
- [Word builder](https://github.com/arjunraj-01/sign-language/blob/main/src/word_builder.py): `src/word_builder.py`
- [Prediction smoother](https://github.com/arjunraj-01/sign-language/blob/main/src/smoothing.py): `src/smoothing.py`

### Live recognition

- [Inference engine](https://github.com/arjunraj-01/sign-language/blob/main/src/inference.py): `src/inference.py`
- [Hand detector](https://github.com/arjunraj-01/sign-language/blob/main/src/hand_detection.py): `src/hand_detection.py`
- [Landmark disambiguation](https://github.com/arjunraj-01/sign-language/blob/main/src/landmark_rules.py): `src/landmark_rules.py`

### Model preparation

- [Dataset loading](https://github.com/arjunraj-01/sign-language/blob/main/src/preprocessing.py): `src/preprocessing.py`
- [CNN model builders](https://github.com/arjunraj-01/sign-language/blob/main/src/model.py): `src/model.py`
- [Model training](https://github.com/arjunraj-01/sign-language/blob/main/train.py): `train.py`
- [Model evaluation](https://github.com/arjunraj-01/sign-language/blob/main/evaluate.py): `evaluate.py`

### Shared settings

- [Runtime configuration](https://github.com/arjunraj-01/sign-language/blob/main/src/config.py): `src/config.py`

### Other components

- **Signer**
- **Webcam**
- **ASL image dataset**
- **Trained classifier**

## Connections

- Signer → Streamlit app: operates (evidence: `README.md`)
- Webcam → Streamlit app: provides frames (evidence: `README.md`)
- Streamlit app → Inference engine: requests predictions (evidence: `app.py`)
- Inference engine → Hand detector: detects hand (evidence: `src/inference.py`)
- Inference engine → Landmark disambiguation: disambiguates (evidence: `src/inference.py`)
- Inference engine → Prediction smoother: feeds predictions
- Streamlit app → Word builder: updates text (evidence: `app.py`)
- Streamlit app → Prediction smoother: stabilizes output (evidence: `app.py`)
- ASL image dataset → Dataset loading: loads images (evidence: `src/preprocessing.py`)
- Model training → Dataset loading: loads splits (evidence: `train.py`)
- Model training → CNN model builders: builds classifier (evidence: `train.py`)
- Model training → Trained classifier: saves weights (evidence: `train.py`)
- Model evaluation → Dataset loading: loads data (evidence: `evaluate.py`)
- Model evaluation → Trained classifier: evaluates
- Model training → Runtime configuration: reads settings (evidence: `train.py`)
- Dataset loading → Runtime configuration: reads settings (evidence: `src/preprocessing.py`)
- Streamlit app → Runtime configuration: uses defaults
