bubbliiiing / fcos-pytorch
这是一个fcos-pytorch的仓库,可以用于训练自己的数据集。
☆24Updated last year
Related projects: ⓘ
- 这是一个yolact-pytorch的库,可用于训练自己的数据集☆70Updated last year
- Wise-IoU: Bounding Box Regression Loss with Dynamic Focusing Mechanism☆60Updated 8 months ago
- 基于YoloV5的一些魔改及相关部署方案☆63Updated 2 years ago
- Pytorch implementation of the 'Slim-neck by GSConv: a lightweight-design for real-time detector architectures'☆188Updated 2 months ago
- 以Swin Transformer作为骨干网络的YoloX目标检测项目☆71Updated 2 years ago
- 这是一个DETR-pytorch的仓库,可以训练自己的数据集☆157Updated last year
- 可以训练yolov5(v6.0)、yolox、小型网络,添加注意力机制☆63Updated 2 years ago
- Multi-backbone, Prune, Quantization, KD☆152Updated 2 years ago
- 知识蒸馏复现相关☆25Updated 2 years ago
- ☆39Updated this week
- CoordAttention_YOLOX(基于CoordAttention坐标注意力机制的改进版YOLOX目标检测平台)☆48Updated last year
- 这是一个efficientnet-yolo3-pytorch的源码,将yolov3的主干特征提取网络修改成了efficientnet☆142Updated 11 months ago
- 这里面存放了一些目标检测算法的数据增强方法。如mosaic、mixup。☆149Updated last year
- ☆106Updated 2 years ago
- ☆68Updated 3 years ago
- ☆77Updated 3 years ago
- 一个修改YOLOv5以使用SwinTransformer模块的代码仓库。A repository that modifies YOLOv5 to use various SwinTransformer blocks.☆100Updated 2 months ago
- YOLOv5的轻量化改进(蜂巢检测项目)☆68Updated 2 years ago
- Make it easier for yolov6 to change the network structure☆67Updated last year
- ☆134Updated 2 years ago
- ☆176Updated 6 months ago
- 🚀🚀🚀YOLOC is Combining different modules to build an different Object detection model.Including YOLOv3、YOLOv4、Scaled_YOLOv4、YOLOv5、YOLO…☆71Updated 2 years ago
- ☆27Updated last year
- ☆79Updated 10 months ago
- https://zhuanlan.zhihu.com/p/430850089☆147Updated last year
- 这是一个yolov5-v6.1-pytorch的源码,可以用于训练自己的模型。☆114Updated last year
- yolov3基础上添加cam与frm模块参考context_augmentation_and_featu☆23Updated 2 years ago
- 本项目支持对剪枝后的yolov5模型进行知识蒸馏训练(This project supports knowledge distillation training for the pruned YOLOv5 model)☆72Updated 7 months ago
- 实现对YOLOX的剪枝操作,添加了卷积层和BN层融合推理,添加中间层可视化功能,可实现预测和训练日志保存☆42Updated last year
- Trans DOTA OBB format(poly format) to YOLO format.☆195Updated 3 years ago