【AI实战营第二期】第四次作业提交05班
作业:MMSeg 语义分割 背景:西瓜瓤、西瓜皮、西瓜籽像素级语义分割
TO DO LIST:
- Labelme 标注语义分割数据集(子豪兄已经帮你完成了)
- 划分训练集和测试集(子豪兄已经帮你完成了)
- Labelme 标注转 Mask 灰度图格式(子豪兄已经帮你完成了)
- 使用 MMSegmentation 算法库,撰写 config 配置文件,训练 PSPNet 语义分割算法
- 提交测试集评估指标
- 自己拍摄西瓜图片和视频,将预测结果发到群里
- (选做)训练 Segformer 语义分割算法,提交测试集评估指标
西瓜瓤、西瓜籽数据集:
标注:同济子豪兄
| 类别名称 | 类别语义 | 标注类别 | 灰度图像素值 |
|---|---|---|---|
| / | 背景 | / | 0 |
| red | 西瓜红瓤 | 多段线(polygon) | 1 |
| green | 西瓜外壳 | 多段线(polygon) | 2 |
| white | 西瓜白皮 | 多段线(polygon) | 3 |
| seed-black | 西瓜黑籽 | 多段线(polygon) | 4 |
| seed-white | 西瓜白籽 | 多段线(polygon) | 5 |
数据集下载链接:
- Labelme标注格式(没有划分训练集和测试集):https://zihao-openmmlab.obs.cn-east-3.myhuaweicloud.com/20230130-mmseg/dataset/watermelon/Watermelon87_Semantic_Seg_Labelme.zip
- Mask标注格式(已划分训练集和测试集):https://zihao-openmmlab.obs.cn-east-3.myhuaweicloud.com/20230130-mmseg/dataset/watermelon/Watermelon87_Semantic_Seg_Mask.zip
需提交的测试集评估指标:(不能低于 baseline 指标的 50% ) aAcc: 60.6200 mIoU: 21.1400 mAcc: 28.4600
We recommend using English or English & Chinese for issues so that we could have broader discussion.
https://github.com/crazy-winds/openmmlab-learn/tree/main/homeworks/Watermelon_seg
https://github.com/ThanosGGGGG/openmmhomework/tree/homework4
https://github.com/hellolele/openmmlab_homework/tree/main
https://github.com/zhouyecs/OpenMMLab-Homework/tree/master/h4
https://github.com/Hust-Liaoyuyang/OpenMMLabHomeWork
https://github.com/x-cosmic/openmmlab_study/tree/main/assignment04
https://github.com/systones/openmmlab-work/blob/main/%E4%BD%9C%E4%B8%9A4.md
https://github.com/ASHORE1225/homework/blob/main/%E4%BD%9C%E4%B8%9A4.pdf
https://github.com/systones/openmmlab-work/blob/main/%E4%BD%9C%E4%B8%9A4.docx 补交详细配置和照片
https://github.com/jercylew/opemmmlab-camp-assignment/tree/main/assignment-4
https://github.com/kakasi1985/openmmlab_homework/tree/main/MMSegmentation_homework