On-Device Real-Time Blurring (SNU Ambient AI — 1st Prize)

Real-time privacy blurring at target FPS on-device under thermal and power constraints — 1st Prize, SNU Ambient AI Competition.

Outcome. Built a mobile blurring pipeline achieving real-time FPS with consistent quality under battery and thermal limits; won 1st prize.

Role: Team member (4) · Dates: Aug–Sep 2024 · Stack: PyTorch-Mobile / NNAPI, Android, OpenCV

Highlights

  • Explored model/runtime trade-offs (quantization, input scaling) with a deterministic harness for latency & quality.
  • Implemented graceful degradation policies to sustain FPS under thermal throttling.
  • Delivered demo app and benchmark report; coordinated split workstreams (model, runtime, UX).
giscus comments misconfigured

Please follow instructions at http://giscus.app and update your giscus configuration.