Skip to content

todo.md

Latest commit

 

History

History
164 lines (95 loc) · 4.61 KB

todo.md

File metadata and controls

164 lines (95 loc) · 4.61 KB

✅ TODO: SnapSort

This document outlines upcoming features, improvements, and bug fixes for SnapSort, organized by priority for efficient tracking and development.

🔁 Version 2.0.0 Plan

🧠 Core Functionality

  • Adding Human Feedback Loop to club same faces that are recognized as different
    A complete different window which contains all the faces and human can select the faces that are same.

🔁 Version 1.0.0 Plan

🧠 Core Functionality

  • Display related images with shared faces
    When an image is selected, show all other images where the same people (faces) also appear together.
    Hint: Use the existing face-image mapping in the database to retrieve relevant results.

Installers

  • Create installers for Windows, Linux, and macOS
    Use PyInstaller or Briefcase to package the application for each platform.
    Hint: Ensure all dependencies are included in the installer.

🔴 High Priority

🔧 Critical Logic & Performance

  • The images are not retrieved from DB but from a list that is in-memory.

    • This is a temporary solution. The images should be retrieved from the database to ensure persistence and reliability.

  • Prevent reprocessing of already indexed images

    • Before processing, check if the image has already been analyzed using SQLite index.

  • Improve recognition accuracy and speed

    • Replace or upgrade current face recognition model.
    • Consider GPU-accelerated libraries such as:
      • insightface
      • onnxruntime
      • face_recognition with CUDA

  • Handle image orientation variations

    • Detect and adjust for rotated or flipped faces during processing.

  • Fix grid layout bug on image load

    • Images currently render in a single column until the window is resized.
    • Manually trigger layout/UI updates after thumbnails are inserted.

  • Prevent crash on "Go Back to Gallery"

    • Handle navigation cleanup correctly to avoid panel memory issues or thread errors.

🟡 Medium Priority

🎨 GUI / UX Improvements

  • Improve overall GUI layout

    • Consistent spacing, padding, and transitions.
    • Polished visual hierarchy for thumbnail views, detail panels, and navigation.

  • Add image interaction options

    • Right-click or hover options for:
      • Delete image
      • Reprocess face data
      • Rename or tag
      • Mark as favorite

  • Add loading/progress indicators

    • Show when thumbnails or face data are being generated in the background.

  • Enable image zoom/preview

    • Click on image to view full-screen or high-res version.

  • Toast notifications or status bar

    • For background events like face matching, DB updates, or errors.

🧠 Functional Enhancements

  • Allow naming of recognized individuals

    • Associate a human-readable name with each face ID via GUI input.

  • Add search functionality

    • Allow searching images by person name or face ID.

  • Show face match confidence score

    • Display the similarity score on hover or click for each matched face.

  • Add pagination or lazy-loading for large datasets

    • Prevent UI freezes and optimize memory usage.

🚀 Deployment & CI/CD

  • Add GitHub Action to build & deploy desktop app
    • Use PyInstaller or Briefcase to package as a desktop app.
    • Auto-generate platform-specific installers (Windows, Linux, macOS).
    • Set up GitHub CI workflow for releases.

🔵 Low Priority

🧰 Code Quality & Architecture

  • Refactor to modular or MVC architecture

    • Improve separation between GUI, logic, and data layers.

  • Add type hints and docstrings

    • Improve readability and help with static analysis.

  • Add configuration file

    • Use config.json, .env, or .ini for user-configurable settings:
      • Paths
      • Face match threshold
      • Cache size

🧪 Testing & DevOps

  • Add unit and integration tests

    • Focus on face recognition logic, database integrity, and UI navigation.

  • Set up GitHub Actions for code quality

    • Run flake8, black, and pytest automatically on push.

  • Maintain dependencies and requirements

    • Keep requirements.txt or pyproject.toml up to date.

✨ Stretch Goals

  • Deploy to mobile (Android/iOS)

    • Consider using Kivy, BeeWare, or React Native with embedded face models.

  • Cloud sync support

    • Optional: Sync metadata or processed DB to a cloud backend for portability.