Show HN: NiceShot AI – The analytics layer competitive games forgot to add
TL;DR
I've been working on NiceShot AI for the last 26 months as a way to experiment with computer vision for gameplay analysis. The basic idea is to take a recorded long gameplay session and automatically turn it into structured gameplay information and highlights. Bridging the performance analysis gap between match stats (too short) & lifetime stats (too long).
Nauti's Take
The modular design is the promising part: supporting a new game should mean a new detector model plus config rather than a rewritten pipeline. The risk sits in detection quality, since YOLO plus OCR on HUD elements breaks when patches, skins or UI updates land.
Coaching teams and serious creators get real value here, while anyone without clean training data should budget for the labeling work first.
Video
Sources