CASE_STUDY // 01 14 Weeks

FoodLens AI

Edge-based Computer Vision & AR Nutritional HUD

ROLE & ARCHITECTURAL LEADERSHIP
Lead Mobile Systems Architect
CLIENT / DOMAIN EXPERTISE
Healthcare Tech & AI Research
VERIFIED_IMPACT_METRICS
Classification Accuracy
90%+
Edge Inference Latency
42ms
Model Footprint
4.8MB
ARCHITECTURAL_CHALLENGE

Modern dietary tracking applications suffer from extreme user friction due to manual logging. The architectural challenge was to build an edge-capable computer vision system that could instantly analyze complex, mixed-food plates in real-time through a mobile camera stream, without causing device thermal throttling or relying entirely on high-latency cloud APIs.

ENGINEERED_SOLUTION

Engineered a custom Flutter camera plugin utilizing native C++ bridges for zero-copy video frame extraction. Integrated an INT8 quantized TensorFlow Lite model directly optimized for Apple CoreML and Android NNAPI execution. For complex ambiguous ingredients, the system intelligently dispatches compressed vector embeddings to Gemini AI as a fallback verification layer, projecting nutritional vectors onto a custom AR Canvas HUD.

CORE_ARCHITECTURE_PILLARS
01. Zero-Copy Frame Extraction

Native C++ camera pipelines bypass standard platform overhead to feed raw byte buffers directly to the NPU.

02. Hybrid Edge-Cloud Inference

Local TFLite models handle 80% of standard classifications in <45ms; edge-case foods trigger asynchronous Gemini multimodal verification.

03. AR Canvas Projection HUD

Custom Flutter rendering engine calculates bounding box matrices and stabilizes UI overlays over moving camera streams.

DEPLOYED_TECHNOLOGY_STACK
FlutterTensorFlow LiteGemini AICoreMLAndroid NNAPIC++ Native Bridge