Keras-Semantic-Segmentation icon indicating copy to clipboard operation
Keras-Semantic-Segmentation copied to clipboard

Keras-Semantic-Segmentation

Results 14 Keras-Semantic-Segmentation issues
Sort by recently updated
recently updated
newest added

hello,这么好的项目,作者不更新了吗?

![image](https://user-images.githubusercontent.com/64268860/111445575-7f694d00-8746-11eb-8c4d-9f4ca673f0a2.png)

这不是真正意义上的SegNet吧,因为保存最大index的那一步好像直接忽略了,直接就是将Unet中Ski-connnection去掉之后的网络吧,应该没看错吧??/

isn't this implementation of ICNet missing the **Cascade Label Guidance**? (section 3.4 in the paper) also i can't see the **weighted softmax cross entropy loss** which should be applied to...

预测图颗粒感严重,请问怎么改善?

我试了一下数据增强,增强的mask是24位的,训练需要8位的mask,需要统一转化为8位的mask吗?

Hi, I'm using PSPNet_ResNet50 model but I got the following error: ValueError: Negative dimension size caused by subtracting 60 from 28 for 'ave_c1_2/AvgPool' (op: 'AvgPool') with input shapes: [?,28,28,2048]. I...

您好,非常感激您的開源分享,你的Github程式非常實用! 還在學習中。 想問我製作自己的數據集(640x480圖片),訓練**ICNet模型**,但執行test.py分割出來的結果"**顆粒感很重**",是正常的嗎? (數據有一張原圖,對應一張黑色的標籤圖(像素0,1,2,3...)存放) ![0](https://user-images.githubusercontent.com/55972088/82830989-c3564180-9ee9-11ea-92ab-1d7285ed8255.jpg) 看到這裡的輸出resize是120x160,這個能改大嗎? o_shape = Model(inputs, o).output_shape outputHeight = o_shape[1] #輸出120 outputWidth = o_shape[2] #輸出160 print("o_shape=",o_shape[1]) print("o_shape=",o_shape[2])

你好,你这个数据生成器是不是没有打乱数据啊,这样的话所有epoch里对应batch中的数据是一样的吧

I need to save a model to pb format in train.py. Do you know how how to do it? Kaishi