UPDF AI

Slim-neck by GSConv: A better design paradigm of detector architectures for autonomous vehicles

Hulin Li,Jun Li,3 Authors,Qiliang Ren

2022 · DOI: 10.48550/arXiv.2206.02424
arXiv.org · 467 Citations

TLDR

This work introduces a new method, GSConv, to lighten the model but maintain the accuracy, and provides a design paradigm, slim-neck, to achieve a higher computational cost-effectiveness of the detectors.