👁️ Assistive Vision Technology
I develop assistive technology for the visually impaired, leveraging the power of on-device machine learning to provide real-time understanding of the world.
Focus Areas
- Real-time Object Detection: Utilizing TensorFlow Lite and CameraX to detect obstacles and objects in real-time.
- Performance Optimization: Focusing on model quantization and efficient inference to ensure low latency and battery efficiency on mobile devices.
- Accessibility: Translating visual data into auditory or haptic feedback to assist users in navigation.
🛠 Technical Stack
On-Device ML
- Inference: TensorFlow Lite, MediaPipe.
- Optimization: Int8 Quantization, Edge Inference.
Android Vision
- Camera: CameraX.
- Acceleration: Neural Networks API (NNAPI).