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

或许数据处理的时候计算iou存在错误 #10

Open
xiaowenhe opened this issue Sep 23, 2020 · 2 comments
Open

或许数据处理的时候计算iou存在错误 #10

xiaowenhe opened this issue Sep 23, 2020 · 2 comments

Comments

@xiaowenhe
Copy link

在dataset.py 第234行计算IOU的时候 iou_scale = self.bbox_iou(bbox_xywh_scaled[i][np.newaxis, :], anchors_xywh),代码中用的anchor是实际的相对与实际坐标的anchor大小,比如【10,13, 16,30, 33,23, 30,61, 62,45, 59,119, 116,90, 156,198, 373,326】,而bbox_xywh_scaled确是除以strides之后的,这样的话,是否存在矛盾呢? 比如,某gt的宽高都为608,训练时的图片最大宽高,假设gt占满全图。则608/8 = 76。也就是说,bbox_xywh_scaled里面的宽高最大才78,那anchor里面的 大于这个宽高的anchor还有什么用呢?所以,这个地方计算iou的时候感觉存在问题。

以上是个人理解,有不对的,请谅解!

@1726251215
Copy link

按照作者给的测试代码和训练代码,对应的anchor应该是[10/8,13/8, 16/8,30/8, 33/8,23/8, 30/16,61/16 ... 373/32, 326/32]

@avBuffer
Copy link
Owner

Hey all, Anchor values maybe caculate according to your train data, and then replace the anchor file.

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

3 participants