Questions tagged [image-segmentation]

The tag has no usage guidance.

54 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
6
votes
0answers
2k views

Why is my Keras model not learning image segmentation?

Edit: as is turns out, not even the model's initial creator could successfully fine-tune it. This is most likely a problem of implementation, or possibly related to the non-intuitive way in which the ...
3
votes
1answer
116 views

Segmentation Network produces noisy output

I've implemented a SegNet and SegNet ReLU variant in PyTorch. I'm using it as a proof-of-concept for now, but what really bothers me is the noise produced by the network. With ADAM I seem to get ...
2
votes
0answers
27 views

Video segmentation vs image segmentation

I am new to data science & working on a segmentation model, Basically I need to deploy this segmentation model in android devices using TensorFlow-Lite for real time camera frame segmentation. I ...
1
vote
1answer
17 views

Photorealistic synthetic data for object segmentation

Let's say that we have a very small labelled dataset for instance segmentation and there is a photorealistic physics engine available that can produce synthetic data for us. Looking on the web I haven'...
1
vote
0answers
73 views

How do you deal with variable input sizes with an encoder-decoder net with skip connections in Keras?

I am currently getting into image segmentation with Keras, and I am using an encoder-decoder type as in the image below. My problem is that applying a MaxPooling2D ...
1
vote
1answer
55 views

Normalization of CT scans

I trained an infection segmentation models on a large dataset of CT scans, and want to extend it to other datasets to show the ability of the model to generalize. What I found though, is that CT scans ...
1
vote
1answer
30 views

Ways of calculating the area of colored regions in a map

Background I am a PHD student trying to improve my data science. One of my research projects, has me tasked with determining the size of the clusters in a colored image of regions. Here is an example ...
1
vote
0answers
13 views

How to approach coarse-grained semantic segmentation?

I’m looking at doing something like semantic segmentation of images but where I only have pretty coarse-grained labels - roughly, for each 32x32 patch, I know if the answer should be “yes”, “no” or “...
1
vote
0answers
751 views

Image Segmentation Class weight using tensorflow keras

I remember definitely being able to pass a list to class_weight with keras (binary image segmentation specifically). For example: ...
1
vote
0answers
76 views

What is the difference between proposal-based approach and proposal-free approach?

From here it says that Techniques to solve instance segmentation can be roughly grouped into two categories: proposal-based methods and proposal-free methods. In proposal-based methods, a set of ...
1
vote
0answers
24 views

Paragraph extraction from text

I am trying to separate scanned pages of a 3 column book into paragraphs. On the pages there can be images located in an arbitrary location, occupying part of one, two or all three of the columns. ...
1
vote
1answer
281 views

How to prepare masks for multiclass semantic segmentation?

It's very straightforward for binary semantic segmentation: black color (0s) is responsible for background, whereas white color (1s) is responsible for objects of interest. But what about multiclass ...
1
vote
0answers
34 views

Image segmentation for small images with rectangles

I am doing toy project and trying to make image segmentation for rectangles generated in small images. Here the code: ...
1
vote
0answers
17 views

An CNN seems like capturing specific range of input data. (Image Segmentation)

I'm trying to build a model to segment brain tumors. I trained a model, and the validation dice coefficient is disappointing(0.6). When i saw the predicted images with the ground truths, it seems ...
1
vote
0answers
15 views

Is there a way to test out simple filters before committing to coding them?

Is there a way to test out simple filters before committing to coding them? Like if I want to estimate the feasibility of recognizing some features from images. Or to estimate the effort/...
1
vote
1answer
17 views

Tools suitable for semi-automatic video labeling?

So far I have been using labelme to label objects in videos I use for training, but it is quite time consuming. Are there good tools to help with that? I was thinking about a tool where I label ...
0
votes
0answers
10 views

How to use mean IoU for RGB mask (keras implementation)?

I am training pix2pix GAN for converting SAR satellite images to segmentation mask. But I am not aware about how to use the mean IoU to evaluate my model. My output is of ...
0
votes
0answers
15 views

Best algorithm for real-time instance segmentation in videos?

I want to do object detection in real-time (meaning localization and classification) on videos (25 FPS), but with the added constraint that my training data is labelled using binary masks, rather than ...
0
votes
0answers
11 views

How can I create .nii (nifti) file from 3D Numpy array

I have a prediction numpy array. How can I make a .nii or .nii.gz mask file from the array?
0
votes
0answers
15 views

U-Net doesn't work with images different from the dataset

I have implemented a very similar U-Net code from github, but for a different dataset, this one, to segment roads, it works fine using the test folder images, but when i for example, pick a print from ...
0
votes
0answers
6 views

Instance segmentation mask creation for adjacent objects within the same class

How are instance segmentation masks created for adjacent objects from the same class ? If for example we have different objects, we can distinguish between them using the labels, but how to ...
0
votes
0answers
7 views

Boundary segmentation

I have one problem regarding segmentation. I would like to put more concentration on the boundary instead of the interior of segmented part. Is that possible to do using Tensorflow 2? For example, if ...
0
votes
1answer
16 views

Semantic segmentation in high-resolution images with high variance - cannot avoid underfitting

I am working on a dataset of 2K images for a semantic segmentation problem. I want to detect and localize small objects, with the smallest mask to be 5x5 pixels. The images include 5 different ...
0
votes
0answers
14 views

Changing order of input dimension in Tensorflow 3-D Layers

According to the official documentation of tf.keras.layers.Conv3D 5+D tensor with shape: batch_shape + (channels, conv_dim1, conv_dim2, conv_dim3) if data_format='channels_first' or 5+D tensor with ...
0
votes
0answers
22 views

Agglomerative Hierarchical Clustering on Images

My goal is to implement the agglomerative hierarchical clustering algorithm on an RGB image to cluster every pixel until some stopping criteria is reached. In order to do so, I assumed that each pixel ...
0
votes
0answers
11 views

Semantic segmentation to instance segmentation

Having a mask from semantic segmentation I want to split it to list of non-overlapping instances masks(like an output from instance segmentation). Do you know some fast algorithm, approach, tool or ...
0
votes
0answers
9 views

I have no access to gpu due to usage limits?

I start running my code using google colab I first set the execution to GPU and then I run my code for a training task using keras !after 1 hour I got a message saying I can't use GPU due to usage ...
0
votes
0answers
55 views

Implementing Multiclass Dice Loss Function

I am doing multi class segmentation using UNet. My input to the model is HxWxC and my output is, ...
0
votes
0answers
12 views

Satellite Change Segmentation using Unet

Hi StackExchange community I am working on to train a Unet for satellite change segmentation. My dataset consists of images(before change),images(after change) and the corresponding change ...
0
votes
0answers
8 views

How to generate fixed number of superpixels?

A lot of work regarding Graph Neural Networks require fixed number of nodes. In the case of image processing using graph, the image representation is often super pixels (like in this work https://...
0
votes
0answers
15 views

Loss function to compare non binary segmentation

I need to compare two images corresponding to landmark locations. I was thinking of something related to Dice loss. I cannot use dice loss since the image is not binary. The background is black but ...
0
votes
0answers
24 views

loss function for multi-label segmentation with class inbalance

In order to use a binary segmentation loss function in a multi label problem, I would like to permute the batch axis with the channel axis in the loss computation in order to compute the loss by ...
0
votes
0answers
9 views

How to Convolve a High-Res Image by a (fully convolutional) CNN kernel?

My CNN is an extremely simple neural network. ...
0
votes
0answers
19 views

IOU and dice_coef exceed 100% in multi task learning

Im doing a multi task learning for road and center line extraction (2 classes) I used IOU and dice_coef as a metrics : ...
0
votes
1answer
336 views

How can I use my own dataset for Image segmentation using Tensorflow

I have a huge problem using my own created dataset for image segmentation using Tensorflow. The dataset that I've build contain images like the one shown below: The problem that I have is: How do I ...
0
votes
0answers
17 views

Model selection Tensorflow for custom dataset comparing panoramic images vs regular images (Image segmentation)

I have a question regarding the use of a specific model such as Deeplab and how to create a custom dataset for it. Background To give a bit of background info, I want to compare panoramically stitched ...
0
votes
0answers
15 views

Why it is reshaped the last layers of VGG_UNet segmentation model?

I want to do a multiclass segmentation task using deep learning (in python). Here, is a summary of vgg_unet model that is mainly collected from GitHub. So, in my dataset 8 labels are available. So, at ...
0
votes
0answers
14 views

Is it possible to create a 3d to 2d U-net?

I was curious if it is possible to create a U-net type architecture that takes in a 3d image and outputs a 2d image? Or, alternatively, would some other architecture be better suited for this problem? ...
0
votes
0answers
45 views

Why does dice coefficient equally weigh false positive and false negative?

According to this article: Dice coefficient= The 2-class DSC variant for class c is expressed in Equation 1, where gic ∈ {0, 1} and pic ∈ [0, 1] represent the ground truth label and the predicted ...
0
votes
0answers
52 views

Image segmentation network to extract questions from an image of a test paper?

This is the sample document -> I want to extract questions along with the options. There are other question papers as which have questions with diagrams in them. I want to be able to extract them ...
0
votes
0answers
16 views

How to determine the decrease of resolution in per cent necessary for a partly blurred frame to appear sharp?

Given is a frame of a video taken with a lens focussing on the background, so the foreground is slightly blurred. How to determine the decrease of resolution in per cent necessary for the blurred ...
0
votes
0answers
18 views

DICE loss too low but no overlap between prediction and label

I am trying to achieve the segmentation of the bone on the cross sectional area of MRI images with the Unet I found here. The label is a binary png image which I intend to compare to my prediction. ...
0
votes
0answers
78 views

How to deal with severe overfitting in a UNet Encoder/Decoder CNN in a task very similar to image translation?

I am trying to fit a UNet CNN to a task very similar to image to image translation. The input to the network is a binary matrix of size (64,256) and the output is of size (64,32). The columns ...
0
votes
0answers
9 views

Segmentation-free character recognition on an image: Multi-label, multi-class or sequential image classification problem?

I have some images which look like this one: They exist of 3 possible characters (A-C) and a length of 4. Now, I would like to run a neural network, which recognizes each character in the picture ...
0
votes
0answers
18 views

How to reduce the detection time in MaskRCNN

I've trained MaskRCNN in GPU instance and using the saved weight, I detected it in CPU instance based system. The detection time is taking too long. ...
0
votes
0answers
6 views

Refining segmentation masks using predictions of earlier frames

I am performing semantic segmentation. For that I have a dataset of videos in which every frame is labeled. Unfortunately, I don't know the order of the frames. That is why I am performing classic ...
0
votes
0answers
355 views

per class IOU & Jaccard Similarity in a Multiclass setting python

For a multiclass classification problem, How do you compute per class IOU ? I am using the formula which is referenced/accepted in the below link ...
0
votes
0answers
10 views

Is it possible to find corresponding coordinates from a downsampled image?

Beforehand, I must apologize, I am just a beginner in image processing and I am not very familiar with many computer vision operations. Here is my problem: I am using deep learning to perform a ...
0
votes
0answers
47 views

Drawing bounding rectangle around the tumor cv2

I am working on a project which predicts that the MRI has tumor or not, now the next step is to draw a bounding rectangle around the tumor. I want to extract the tumor from the MRI and bound the ...
0
votes
1answer
73 views

How to save multi-output predicted masks into two different folders after using model.predict_generator

I have a multi output segmentation task, the training process went well, but when Im trying to get the prediction I found difficulties to separate the two output into two different folders, in my ...