Closed CarryHJR closed 4 months ago
个人推测 可能作者论文实验里面的划分 和 上传权重对应的划分不一致,论文里面在FLIR上 mAP50是 79.2
我们一直在优化代码框架,所以目前的实际性能与论文会有所不同。但是融合模块的代码与论文中提出的方法仍然是保持一致的。 We continuously optimize our codes, which results in the difference in detection performance. However, the codes of module for multimodal feature fusion still remain consistent with the methods proposed in this paper.
直接运行
python test.py --weights weights/ICAFusion_FLIR.pt --device 1