@inproceedings{li2019gradient,
title={Gradient Harmonized Single-stage Detector},
author={Li, Buyu and Liu, Yu and Wang, Xiaogang},
booktitle={AAAI Conference on Artificial Intelligence},
year={2019}
}
| Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | Config | Download |
|---|---|---|---|---|---|---|---|
| R-50-FPN | pytorch | 1x | 4.0 | 3.3 | 37.0 | config | model | log |
| R-101-FPN | pytorch | 1x | 6.0 | 4.4 | 39.1 | config | model | log |
| X-101-32x4d-FPN | pytorch | 1x | 7.2 | 5.1 | 40.7 | config | model | log |
| X-101-64x4d-FPN | pytorch | 1x | 10.3 | 5.2 | 41.4 | config | model | log |