1：参考GitHub - PaddlePaddle/PaddleDetection: Object Detection toolkit based on PaddlePaddle. It supports object detection, instance segmentation, multiple object tracking and real-time multi-person keypoint detection.

2：开源项目 - 飞桨AI Studio星河社区-人工智能学习与实训社区

3：训练阶段首先利用原有的数据，然后更改成自己的数据

VOC:数据集满足条件

trian.txt valid.txt

3.2 配置yml  主要 然后根据这个 去更改其它的

cat /workspace/PaddleDetection/configs/ppyoloe/ppyoloe_plus_crn_s_80e_voc.yml

_BASE_: [
  '../datasets/roadsign_voc.yml',
  '../runtime.yml',
  './_base_/optimizer_80e.yml',
  './_base_/ppyoloe_plus_crn.yml',
  './_base_/ppyoloe_plus_reader.yml',
]

log_iter: 100
snapshot_epoch: 5
weights: output/ppyoloe_plus_crn_s_80e_coco/model_final

pretrain_weights: https://bj.bcebos.com/v1/paddledet/models/pretrained/ppyoloe_crn_s_obj365_pretrained.pdparams
depth_mult: 0.33
width_mult: 0.50

旋转数据集制作

旋转目标检测-环境配置-数据集制作_旋转目标检测数据集-CSDN博客

工具是 GitHub - cgvict/roLabelImg: Label Rotated Rect On Images for training