Michael Chen

Project

PHiLIP

human-in-the-loop image generation on AMD ROCm

team writeup on Hackster

Part of ML & Research.

Our PHiLIP prototype combines human feedback with image generation using AMD ROCm and PixArt-alpha. It won the AMD University Program Award in the AMD Pervasive AI Developer Contest.

The team's 2024 writeup reports about five seconds for nine guidance images and about five seconds for enhancement. The public source contains 71 style presets.

These are historical prototype timings. The current README documents route and configuration mismatches, plus machine-specific dependency paths. It does not verify model loading or end-to-end operation.

guidance images
~5s per 9
enhancement
~5s
style presets
71

The team writeup describes the prototype and its historical timing. Official contest results confirm the award. The repository README documents the code and its current limits. Style preset source.