Back to Case Studies
Industrial AutomationSmart HomeAgTech
StyleTech
Retail Smart Mirror based on MediaTek MT8183
AR mirror for virtual clothing try-on.
Vision FPS
30 FPS
Thermal Max
45°C
StyleTech
Product View
System Architecture
Client
A high-end fashion retailer requiring interactive fitting room mirrors that suggest accessories based on the garments being tried on.
Challenge
Integrating a large 32-inch capacitive touch screen behind a two-way mirror, utilizing RFID and visual AI to identify clothing items seamlessly.
Product Concept & Architecture
We utilized the MediaTek MT8183 (Octa-core, Mali-G72 GPU) on a custom System-on-Module (SoM). It interfaces with an Impinj RFID reader module via UART and a 1080p camera via MIPI-CSI.
Engineering Solution
We engineered the Android 10 firmware to support large-format touch drivers through glass. When a customer brings clothes into the fitting room, the RFID reader instantly identifies the SKUs. Concurrently, the camera uses a lightweight TensorFlow Lite model to analyze the customer's skin tone and style preference, overlaying matching accessory recommendations on the UI.
Result
Increased accessory upselling by 22% in pilot stores. Mirror touch response feels instantaneous despite the thick two-way glass layer.
Project Profile
Services Delivered
OS PortingHardware Design
Core Technologies
Android AOSPEmbedded LinuxNPU Acceleration
Silicon & Vendors
MediaTekRockchipQuectelAllwinner
Need similar outcomes?
Let's discuss how ChipTalk can deliver results for your next project.
Tell us about your project.
Fast Response — Initial feedback within 48 hours
Full R&D Lifecycle — From feasibility to mass production
Extensive Experience — 500+ projects across 15 verticals
Rapid MVP Validation — Prototype in as fast as 12 weeks
Portfolio
