📡 ForgetMeNot AI™ — Capture Screen

Capture the Signal. Analyze the Context. Preserve the Memory.

The Capture screen is the multimodal input gateway of ForgetMeNot AI™.

It allows the user to introduce a real-world signal into the application through different forms of input. The captured information is then processed through Gemini AI analysis, transformed into structured contextual information, and stored in the user's Firebase database.

The Capture architecture has two independent dimensions:

🎛️ Three Capture Methods

1. 📝 Note / Text
2. 🖼️ Image
3. 🎙️ Voice-Over — Coming Soon

🏷️ Three Signal Tags

1. 👤 People
2. 📍 Places
3. 📦 Things

Any capture method can be associated with any of the three signal tags.

---

📸 Capture Screen

The Capture screen gives the user a direct interface for introducing contextual information into ForgetMeNot AI.

The user can provide a written signal, an image, or, in the planned future release, a voice signal.

The captured information then enters the Gemini intelligence pipeline.

USER
  │
  ▼
CAPTURE
  │
  ├── 📝 NOTE / TEXT
  ├── 🖼️ IMAGE
  └── 🎙️ VOICE — COMING SOON
  │
  ▼
SIGNAL TAG
  │
  ├── 👤 PEOPLE
  ├── 📍 PLACES
  └── 📦 THINGS
  │
  ▼
GEMINI AI ANALYSIS
  │
  ▼
STRUCTURED SIGNAL
  │
  ▼
FIREBASE DATABASE
  │
  ▼
MEMORY

---

🧩 Capture Method vs. Signal Type

It is important to distinguish between how the user provides information and what the information is about.

Capture Method = How the signal enters the application

Method	Description	Status
📝 Note / Text	User types a signal manually	Available
🖼️ Image	User provides a visual signal	Available
🎙️ Voice-Over	User provides a spoken signal	Coming Soon

Signal Tag = What the signal represents

Tag	Description
👤 PEOPLE	Person or relationship-related context
📍 PLACES	Location or destination-related context
📦 THINGS	Object, task, activity, or other contextual information

Therefore:

                 CAPTURE METHOD
                       │
        ┌──────────────┼──────────────┐
        │              │              │
        ▼              ▼              ▼
    📝 TEXT         🖼️ IMAGE       🎙️ VOICE
        │              │              │
        └──────────────┼──────────────┘
                       │
                       ▼
                  SIGNAL TAG
                       │
        ┌──────────────┼──────────────┐
        │              │              │
        ▼              ▼              ▼
    👤 PEOPLE       📍 PLACES      📦 THINGS

A note can therefore be about a person, place, or thing.

An image can also represent a person, place, or thing.

The planned Voice-Over capability will similarly support People, Places, and Things.

---

📝 Capture Method 1 — Note / Text

The first Capture method allows the user to manually enter a note or text signal.

The information may be a simple observation, reminder, task, intention, event, person-related item, location, object, or other contextual information.

Example

"I need to take my passport when I leave for the airport tomorrow."

The user's original text is the raw signal.

At this point, the application has the user's statement, but the intelligence layer has not yet converted it into its structured contextual representation.

RAW USER INPUT
      ↓
"I need to take my passport
when I leave for the airport tomorrow."
      ↓
GEMINI ANALYSIS

---

🖼️ Capture Method 2 — Image

The second Capture method allows the user to provide an image.

The image becomes a multimodal signal that can contain contextual information that is not explicitly typed by the user.

For example, an image could represent:

• 👤 a person
• 📍 a location
• 📦 an object
• a document
• an item that needs to be remembered
• visual contextual information

The image enters the Gemini multimodal analysis pipeline.

IMAGE
  ↓
GEMINI VISUAL / CONTEXTUAL ANALYSIS
  ↓
STRUCTURED SIGNAL

---

🎙️ Capture Method 3 — Voice-Over

Status: Coming Soon

Voice-Over is planned as the third Capture method.

The purpose is to allow a user to communicate a signal naturally through speech rather than typing it or providing an image.

The planned model is:

🎙️ USER VOICE
      ↓
VOICE SIGNAL
      ↓
GEMINI ANALYSIS
      ↓
STRUCTURED CONTEXT
      ↓
FIREBASE

Voice-Over will be designed around the same three signal categories:

🎙️ VOICE
   │
   ├── 👤 PEOPLE
   ├── 📍 PLACES
   └── 📦 THINGS

---

🏷️ The Three Signal Tags

👤 PEOPLE

The People tag represents information connected with a person or people.

Examples may include:

• Someone the user needs to contact
• Someone the user needs to meet
• A person connected with a task
• A person associated with an event
• A person the user wants to remember

A signal can be captured as text, image, or eventually Voice-Over.

---

📍 PLACES

The Places tag represents location-related information.

Examples may include:

• A destination
• Meeting location
• Office
• Airport
• Restaurant
• Travel destination
• Place associated with an upcoming activity

Again, the signal can originate from text, image, or Voice-Over.

---

📦 THINGS

The Things tag represents objects, tasks, activities, belongings, or other contextual items.

Examples may include:

• Passport
• Keys
• Documents
• Shopping items
• Equipment
• An unfinished task
• Something that needs to be taken, prepared, completed, or remembered

---

🧠 Gemini AI Analysis

The Gemini AI analysis layer is the central intelligence stage between the user's raw Capture input and persistent Firebase storage.

The user does not simply store an unprocessed note or image.

The captured signal first enters the AI analysis process.

              USER CAPTURE
                   │
                   ▼
        ┌─────────────────────┐
        │    RAW SIGNAL       │
        │                     │
        │ 📝 Text             │
        │ 🖼️ Image            │
        │ 🎙️ Voice — Soon     │
        └──────────┬──────────┘
                   │
                   ▼
        ┌─────────────────────┐
        │    GEMINI AI        │
        │      ANALYSIS       │
        └──────────┬──────────┘
                   │
                   ▼
        ┌─────────────────────┐
        │ CONTEXTUAL SIGNAL   │
        │     STRUCTURE       │
        └──────────┬──────────┘
                   │
                   ▼
        ┌─────────────────────┐
        │    FIREBASE DB      │
        └─────────────────────┘

Gemini provides the application's interpretation layer for the captured information.

---

🔍 What Happens During Gemini Analysis?

The exact output depends on the information provided by the user, but conceptually the analysis transforms the raw signal into a more structured contextual representation.

The analysis can involve understanding elements such as:

1. Signal Meaning

What is the user communicating?

"I need to take my passport tomorrow."

The system can interpret the statement as information concerning an item that the user needs to remember.

---

2. Signal Category

The information can be associated with the application's signal taxonomy:

📦 THINGS

because the signal concerns a physical item.

---

3. Context

The analysis can identify the surrounding contextual information contained within the user's signal.

For example:

THING
→ Passport

ACTION
→ Take / carry

TEMPORAL CONTEXT
→ Tomorrow

USER INTENTION
→ Remember to take the item

---

4. Relevant Relationships

The signal may contain relationships between different contextual elements.

For example:

PASSPORT
   │
   ├── belongs to → THINGS
   ├── action → TAKE
   └── context → LEAVING / TRAVEL

---

5. Structured Signal

The raw information can then be represented as a structured ForgetMeNot AI signal.

Conceptually:

RAW SIGNAL
    ↓
UNDERSTAND
    ↓
CLASSIFY
    ↓
EXTRACT CONTEXT
    ↓
STRUCTURE
    ↓
STORE

The important transition is:

Raw human input → AI-understood contextual signal

---

🧠 Example — Note → Gemini Analysis

Suppose the user enters:

"I need to take my passport when I leave
for the airport tomorrow."

The Capture pipeline can be represented as:

┌─────────────────────────────────────┐
│ RAW USER NOTE                       │
│                                     │
│ "I need to take my passport         │
│  when I leave for the airport       │
│  tomorrow."                         │
└──────────────────┬──────────────────┘
                   ↓
             GEMINI AI
                   ↓
┌─────────────────────────────────────┐
│ ANALYZED CONTEXT                    │
│                                     │
│ Signal: Passport                    │
│ Category: THINGS                    │
│ Action: Take / Carry                │
│ Context: Airport / Travel           │
│ Time Context: Tomorrow              │
│ Intent: Remember before departure   │
└──────────────────┬──────────────────┘
                   ↓
            STRUCTURED SIGNAL
                   ↓
             FIREBASE DB

This illustrates how ForgetMeNot AI moves from a simple human statement toward a structured contextual memory.

---

🖼️ Example — Image → Gemini Analysis

An image can follow the same overall architecture.

┌──────────────────────────┐
│          IMAGE           │
└────────────┬─────────────┘
             ↓
┌──────────────────────────┐
│       GEMINI AI          │
│   MULTIMODAL ANALYSIS    │
└────────────┬─────────────┘
             ↓
┌──────────────────────────┐
│ CONTEXTUAL INTERPRETATION│
│                          │
│ What is visible?         │
│ What does it represent?  │
│ Which signal category?   │
│ What context is relevant?│
└────────────┬─────────────┘
             ↓
┌──────────────────────────┐
│   STRUCTURED SIGNAL      │
└────────────┬─────────────┘
             ↓
┌──────────────────────────┐
│      FIREBASE DB         │
└──────────────────────────┘

The image therefore becomes part of the same persistent signal architecture as a text-based capture.

---

📸 The Three Wiki Screenshots

The Capture page uses three screenshots to demonstrate representative states of the Capture functionality.

These screenshots should not be interpreted as three separate signal categories.

They demonstrate different Capture / Analysis states.

---

Screenshot 1 — Note Before Gemini Analysis

This screenshot represents the user providing a Note / Text signal before Gemini analysis.

At this stage:

USER
 ↓
NOTE / TEXT
 ↓
RAW SIGNAL
 ↓
WAITING FOR ANALYSIS

The information shown represents the user's original input.

---

Screenshot 2 — Note With Gemini Analysis

This screenshot represents the same general Capture workflow after the signal has gone through Gemini AI analysis.

The application can now work with the analyzed contextual representation rather than only the user's original text.

NOTE / TEXT
      ↓
GEMINI AI
      ↓
ANALYSIS
      ↓
CONTEXT
      ↓
STRUCTURED SIGNAL
      ↓
FIREBASE

This is the key transition from capture to intelligent memory.

---

Screenshot 3 — Image With Gemini Analysis

This screenshot demonstrates an image-based signal after Gemini analysis.

The image is treated as a multimodal input and processed through the same intelligence architecture.

IMAGE
  ↓
GEMINI MULTIMODAL ANALYSIS
  ↓
CONTEXTUAL UNDERSTANDING
  ↓
STRUCTURED SIGNAL
  ↓
FIREBASE

The Capture architecture therefore supports both textual and visual signals while maintaining the same underlying storage and memory lifecycle.

---

☁️ Firebase Database Storage

After Gemini analysis, the analyzed signal is stored in the Firebase database for the authenticated user.

The complete storage pipeline is:

USER
 ↓
CAPTURE
 ↓
GEMINI AI ANALYSIS
 ↓
STRUCTURED SIGNAL
 ↓
FIREBASE DATABASE
 ↓
USER MEMORY

Firebase becomes the persistent layer containing the user's analyzed signals.

The stored information can subsequently participate in the broader ForgetMeNot AI contextual memory system.

---

🧠 Capture → Gemini → Firebase → Memory

The relationship between the Capture and Memory screens is fundamental to the application.

┌─────────────────┐
│     CAPTURE     │
│                 │
│ Note / Image /  │
│ Voice — Soon    │
└────────┬────────┘
         ↓
┌─────────────────┐
│    GEMINI AI    │
│    ANALYSIS     │
└────────┬────────┘
         ↓
┌─────────────────┐
│    STRUCTURED   │
│     SIGNAL      │
└────────┬────────┘
         ↓
┌─────────────────┐
│   FIREBASE DB   │
└────────┬────────┘
         ↓
┌─────────────────┐
│     MEMORY      │
│                 │
│ View / Edit /   │
│ Delete          │
└─────────────────┘

---

✏️ User Control Through Memory

Once the analyzed signal has been stored in Firebase, the user can access it through the Memory screen.

The user can:

👁️ View

Review the stored contextual signal.

✏️ Edit

Modify information associated with the stored signal.

🗑️ Delete

Remove a stored signal from the user's memory collection.

Therefore, the Capture process does not end with Firebase storage.

It continues into a user-controlled memory lifecycle:

CAPTURE
   ↓
ANALYSIS
   ↓
STORAGE
   ↓
MEMORY
   ├── VIEW
   ├── EDIT
   └── DELETE

---

🔄 Complete ForgetMeNot Capture Lifecycle

                         USER
                          │
                          ▼
                 ┌────────────────┐
                 │     CAPTURE     │
                 └───────┬────────┘
                         │
          ┌──────────────┼──────────────┐
          ▼              ▼              ▼
     📝 NOTE/TEXT    🖼️ IMAGE      🎙️ VOICE
                                    COMING SOON
          │              │              │
          └──────────────┼──────────────┘
                         ▼
                  ┌──────────────┐
                  │ SIGNAL TAG   │
                  └──────┬───────┘
                         │
             ┌───────────┼───────────┐
             ▼           ▼           ▼
         👤 PEOPLE    📍 PLACES   📦 THINGS
             │           │           │
             └───────────┼───────────┘
                         ▼
                  ┌──────────────┐
                  │   GEMINI AI  │
                  │   ANALYSIS   │
                  └──────┬───────┘
                         ▼
                  ┌──────────────┐
                  │  STRUCTURED  │
                  │    SIGNAL    │
                  └──────┬───────┘
                         ▼
                  ┌──────────────┐
                  │  FIREBASE DB │
                  └──────┬───────┘
                         ▼
                  ┌──────────────┐
                  │    MEMORY    │
                  └──────┬───────┘
                         │
                 ┌───────┼───────┐
                 ▼       ▼       ▼
              👁️ VIEW  ✏️ EDIT  🗑️ DELETE

---

📊 Capture Architecture at a Glance

Layer	Function
📝 Note / Text	User enters a written signal
🖼️ Image	User provides a visual signal
🎙️ Voice-Over	Planned spoken signal input
👤 People	Person-related signal classification
📍 Places	Location-related signal classification
📦 Things	Object/task/context signal classification
🧠 Gemini AI	Analyzes the captured signal
🔍 Context Analysis	Interprets relevant information within the signal
🧩 Structured Signal	Converts analyzed information into application-ready context
☁️ Firebase DB	Stores analyzed signals for the authenticated user
🧠 Memory	Displays stored signals
✏️ Edit	User can modify a stored signal
🗑️ Delete	User can remove a stored signal

---

🌌 The Intelligence Behind Capture

The Capture screen is more than a traditional note-taking interface.

It establishes the first connection between:

Human observation → Multimodal input → AI interpretation → Contextual memory

The user provides the information in the most natural available form.

The application then moves that information through the intelligence pipeline:

Capture → Analyze → Structure → Store → Remember

With three capture methods:

📝 Text · 🖼️ Image · 🎙️ Voice

and three signal dimensions:

👤 People · 📍 Places · 📦 Things

the Capture screen forms the foundation of the ForgetMeNot AI predictive memory architecture.

---

💫 Core Principle

The user creates the signal. Gemini analyzes the signal. Firebase preserves the signal. Memory gives the user control over the signal.

ForgetMeNot AI™

From Multimodal Perception to Predictive Preparedness.

Predict the Omission. Prepare Before the Ripple.