Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[New Feature] MM Grounding DINO #11313

Open
hhaAndroid opened this issue Dec 26, 2023 · 8 comments
Open

[New Feature] MM Grounding DINO #11313

hhaAndroid opened this issue Dec 26, 2023 · 8 comments
Assignees

Comments

@hhaAndroid
Copy link
Collaborator

https://github.com/open-mmlab/mmdetection/blob/dev-3.x/configs/mm_grounding_dino/README.md

@hhaAndroid
Copy link
Collaborator Author

#11321

@hhaAndroid
Copy link
Collaborator Author

#11326

@hhaAndroid
Copy link
Collaborator Author

#11322

@hhaAndroid
Copy link
Collaborator Author

#11342

@wu33learn
Copy link

#11342请请问训练自己数据集时,包含多个类别,报错,[data['category_id'] = self.cat_ids[label] IndexError: list index out of range]如何解决,另外,我训练好几轮,一直都是第一个epoch效果最好,但AP,AR都是-1。感谢!

@xuans29
Copy link

xuans29 commented Oct 27, 2024

#11342请请问训练自己数据集时,包含多个类别,报错,[data['category_id'] = self.cat_ids[label] IndexError: list index out of range]如何解决,另外,我训练好几轮,一直都是第一个epoch效果最好,但AP,AR都是-1。感谢!
您好,请问您解决这个问题了吗,我也出现了这个情况

@wu33learn
Copy link

wu33learn commented Oct 27, 2024 via email

@librety123z
Copy link

嗯,你看你数据集对不对,不能有subclass啥的,应该是数据集的问题,你找找mmdetection的文档有指定的标准数据集格式。他这个代码应该是都没错的,我是改了数据集json文件就好了。

吴国成
@.***

郑州大学
学生

请问您使用的是coco数据集格式还是ovdg数据集格式呢?我两种数据集格式都有使用,但是仍然出现这样的问题,我尝试使用了以下几种格式进行训练:
1、参考coco标注:
{
"images": [
{
"height": 4096,
"width": 2896,
"id": 0,
"file_name": "0002-2.png"
},
{
"height": 4096,
"width": 2896,
"id": 1,
"file_name": "0001-1.png"
}
// 更多图像...
],
"annotations": [
{
"iscrowd": 0,
"category_id": 0,
"id": 0,
"image_id": 0,
"bbox": [1818, 734, 1994, 797],
"area": 3792,
"segmentation": []
},
{
"iscrowd": 0,
"category_id": 0,
"id": 1,
"image_id": 0,
"bbox": [1826, 1519, 2202, 1578],
"area": 1296,
"segmentation": []
},
{
"iscrowd": 0,
"category_id": 0,
"id": 2,
"image_id": 0,
"bbox": [1777, 2288, 1889, 2558],
"area": 76710,
"segmentation": []
}
// 更多标注...
],
"categories": [
{
"id": 0,
"name": "cat"
},
{
"id": 1,
"name": "dog"
},
{
"id": 2,
"name": "mouse"
}
]
}
2、参考ovdg格式:
{
"filename": "20002-2.png",
"height": 4096,
"width": 2896,
"detection": {
"instances": [
{"bbox": [1818, 734, 1994, 797], "label": 0, "category": "cat"},
{"bbox": [1826, 1519, 2202, 1578], "label": 1, "category": "dog"},
{"bbox": [1777, 2288, 1889, 2558], "label": 2, "category": "mouse"}
// 更多实例...
]
}
}
{
"filename": "10001-1.png",
"height": 4096,
"width": 2896,
"detection": {
"instances": [
{"bbox": [1818, 734, 1994, 797], "label": 0, "category": "cat"},
{"bbox": [1826, 1519, 2202, 1578], "label": 1, "category": "dog"},
{"bbox": [1776, 2288, 1894, 2558], "label": 2, "category": "mouse"}
// 更多实例...
]
}
}
// 更多图像...
3、参考https://github.com/open-mmlab/mmdetection/blob/main/configs/mm_grounding_dino/usage_zh-CN.md链接中的cat数据集格式:
https://download.openmmlab.com/mmyolo/data/cat_dataset.zip
{
"images": [
{
"height": 4096,
"width": 2896,
"id": 0,
"file_name": "0002-2.png"
},
{
"height": 4096,
"width": 2896,
"id": 1,
"file_name": "0001-1.png"
}
// 更多图像...
],
"categories": [
{
"id": 0,
"name": "cat"
},
{
"id": 1,
"name": "dog"
},
{
"id": 2,
"name": "mouse"
}
],
"annotations": [
{
"iscrowd": 0,
"category_id": 0,
"id": 0,
"image_id": 0,
"bbox": [1818, 734, 1994, 797],
"area": 3792,
"segmentation": []
},
{
"iscrowd": 0,
"category_id": 0,
"id": 1,
"image_id": 0,
"bbox": [1826, 1519, 2202, 1578],
"area": 1296,
"segmentation": []
},
{
"iscrowd": 0,
"category_id": 0,
"id": 2,
"image_id": 0,
"bbox": [1777, 2288, 1889, 2558],
"area": 76710,
"segmentation": []
}
// 更多标注...
]
}

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants