Stack Overflow
Stack Overflow

Questions tagged [semantic-segmentation]

Semantic segmentation is the task of classifying every pixel in the image to a category or class label.

Semantic segmentation is the task of classifying every pixel in the image to a category. It refers to the process of linking each pixel in an image to a class label. Some of its primary applications are in autonomous vehicles, human-computer interaction & robotics.

532 questions
2
votes
0 answers

Binary semantic segmentation with sigmoid layer and tf.keras.metrics.MeanIoU

My network for binary semantic segmentation has a sigmoid activation in the last layer, so all predictions are scaled between 0-1. I want to use the metric tf.keras.metrics.MeanIoU(num_classes) which compares classified predictions (0 or 1) with…
Tiz
  • 41
  • 4
2
votes
0 answers

How to fit data for augmentation to avoid out of memory error?

Doing augmentation for training segmentation but the total number of images is about 26,000+. That's the reason facing problem in making an array of images. Tried: def get_data(): X_data = [] train_images =…
MSI
  • 105
  • 9
2
votes
1 answer

cannot import name 'get_config' from 'tensorflow.python.eager.context'?

I'm trying to follow this repo's tutorial on colabhttps://github.com/divamgupta/image-segmentation-keras but I'm getting this error again and again cannot import name 'get_config' from 'tensorflow.python.eager.context'…
z2ouu
  • 41
  • 1
  • 10
2
votes
1 answer

Loss function for binary classification with problem of data imbalance

I try to segment of multiple sclerosis lesions in MR images using deep convolutional neural networks with keras. In this task, each voxel must be classified, either as a lesion voxel or healthy voxel. The challenge of this task is data imbalance…
NahidEbrahimian
  • 61
  • 1
  • 7
2
votes
1 answer

Problem with training U-Net while using ImageDataGenerator on multiclass segmentation

The task I am dealing with is multiclass segmentation (0-3 classes on each image). I had a working U-Net model and could train on small dataset just fine, then I augmented the dataset and now I have almost 15k 512x512 grayscale images. I've…
Nuwanda
  • 51
  • 1
  • 4
2
votes
1 answer

Convert coco to labelme format

I want to convert my existing coco format into the labelme…
honeymoon
  • 2,400
  • 5
  • 34
  • 43
2
votes
1 answer

Tensorflow (Keras) U-Net Segmentation training fails

I am trying to train a U-Net with Tensorflow and Keras. Model shown below. def get_unet(IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS): inputs = Input((IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS)) conv1 = Conv2D(32, 3, activation='relu', padding='same',…
manuel
  • 1,127
  • 1
  • 8
  • 15
2
votes
2 answers

Semantic Segmentation on a bigger image

I trained a U-net with inputs of satellite images of 120 X 120. I need to apply my model to a bigger image (size 10980 X 10980). What I tried to do was slice the bigger images into slices of 120 X120 classify these and assemble them into a new…
Gon
  • 51
  • 5
2
votes
0 answers

Pytorch's Equivalent of keras train_on_batch()

acc_loss = model.train_on_batch([support, smask, query], qmask) This is a part of code that I want to convert to pytorch but I'm stuck to find out if there's any Pytorch's function as train_on_batch() or any custom function in Pytorch that can…
Codess
  • 21
  • 2
2
votes
2 answers

Why is the predicted output (value) from segmentation model library (0 to 1) instead of (0 or 1)?

While performing semantic-segmentation task by following this tutorial , I noticed that the final predicted output from the model is not 0 and 1, it consists of decimal values from 0.0000xxxx to 1.0. Since the model took in the label of 0 and 1…
jona
  • 592
  • 1
  • 4
  • 16
2
votes
2 answers

Evaluate U-Net by layer

I am coming from medical background and a newbie in this machine learning field. I am trying to train my U-Net model using keras and tensorflow for image segmentation. However, my loss value is all NaN and the prediction is all black. I would like…
jonedabb
  • 59
  • 6
2
votes
1 answer

Dice coefficient for image segmentation evaluation

I'm trying to implement dice coefficient, so I could compare segmented image to ground truth image. However, the output of the dice coefficient seems to be incorrect as segmented image is around 80% similar to the ground truth image. The output of…
Tyler Durden
  • 31
  • 4
2
votes
1 answer

Resizing a target segmentation map without converting individual pixel values to floats

I have a dataset with drone view images of size 4000x6000, grayscale. Each individual pixel value corresponds to a class (I have 20 classes in total), so a pixel value of 3 would mean "tree" for example. Using the original image, I can very easily…
Aditya Mehrotra
  • 323
  • 1
  • 18
2
votes
0 answers

How is the smooth Dice coefficient is differentiable but not the smooth IoU?

I would like to know how the smooth Dice coefficient is differentiable but not a smooth IoU. Exactly which part of smooth IoU makes is non-differentiable, since the equation for two seems quite similar. Calculus or differentiation isn't my strong…
Roman
  • 31
  • 4
2
votes
2 answers

Why is dilated convolution computationally efficient?

I have been studying UNet inspired architecture ENet and I think I follow the basic concepts. The ground-rock of efficiency of ENet is dilated convolution (apart other things). I understand the preserving spatial resolution, how it is computed and…
Michal Cicatka
  • 55
  • 1
  • 7
1 2 3
…
35 36