DeepLab is a deep learning model for semantic image segmentation, where the goal is to assign semantic labels to every pixel in the input image.
Questions tagged [deeplab]
130 questions
1
vote
1 answer
Tensorflow mIOU and pixel accuracy bug?
Let's say I started training a tensorflow model from scratch with 1000 training steps. I get the following result at the completion of training.
Now, lets say I want to train for 2000 training steps from the previously saved checkpoint. I get the…
Simran Marok
- 707
- 1
- 5
- 9
1
vote
1 answer
1
vote
1 answer
Why is the deeplab v3+ model confused about pixels outside image boundary?
I'm using the google research github repository to run deeplab v3+ on my dataset to segment parts of a car. The crop size I've used is 513,513 (default) and the code adds a boundary to images smaller than that size (correct me if I'm wrong).…
rick_moody
- 50
- 9
1
vote
1 answer
Tensorflow: TF_SessionRun returns TF_INVALID_ARGUMENT
I am running a session from a frozen graph of Deeplabv3 using the Tensorflow C API. When I get to the part of running the session with TF_SessionRun, the return value is 3, indicating TF_INVALID_ARGUMENT. I suspect it may have to do something with…
Carlos Vazquez
- 426
- 3
- 9
1
vote
1 answer
My custom mobilenet trained model is not showing any results. What am I doing wrong?
I started to learn ML using Tensorflow/Deeplab. I tried to train my own model from scratch for clothes recognition using semantic segmentation with mobilenet_v2 model variant. But I don't get results.
I'm using tensorflow/models for tfrecord export…
Brayan Bonilla
- 13
- 4
1
vote
1 answer
There was a problem running the test text 'local_test.sh' in deeplab.But modle_test.py works fine
I copied deeplab's source code on github and configured all the files as required.And it works fine for model_test.py.But when I tried to run the local_tset.sh test file, a series of problems occurred.
I can't read the error message, so I don't know…
z korpse
- 11
- 1
1
vote
1 answer
How to fix 'Invalid argument: Got 3 frames, but animated gifs can only be decoded by tf.image.decode_gif or tf.image.decode_image'
I am using the deeplab framework in order to classify images (Github: https://github.com/tensorflow/models/tree/master/research/deeplab with tensorflow version 1.14.0) that have more than 4 channels of information.
I had the idea to put separate…
Michael Scott
- 11
- 3
1
vote
1 answer
Problem with the visualization output from Deeplab v3+
I am trying to run Deeplab v3+ (the standard tensorflow version) on some remote sensing data to perform a binary classification (hedge or no hedge), but I am finding the output to be very strange, leading me to believe that there may be something…
Steve Ahlswede
- 547
- 1
- 4
- 14
1
vote
1 answer
Why is xception_71 inference time faster than that of xception_65 in deeplab?
I'm doing inference for semantic segmentation using pretrained (on cityscapes dataset) DeepLab models based on xception_65 and xception_71 architectures. I've observed that:
xception_65 is better in creating segmentation masks compared to…
kHarshit
- 11,362
- 10
- 52
- 71
1
vote
0 answers
LibTorch, using deeplab model gives segfault on forwarding
I am trying to run images through the DeepLab model in Libtorch to segment them. Using pytorch, i am converting the Deeplabv3 model like this:
import torch
import torchvision
from torchvision import models
deeplap_model =…
anti
- 3,011
- 7
- 36
- 86
1
vote
0 answers
having problems migrating from DeepLab to DeepLabV3 in iOS
I am currently using a 'DeepLab-V1' to do image segmentation in an iOS native app, i want to migrate to 'DeepLab-V3' but can seem to find a way to do so.
Any example of 'DeepLab-V3' implemented in a Xcode project would be appreciated or…
Kenpachi
- 11
- 1
1
vote
0 answers
AttributeError: dense_prediction_cell_json - Why can't the json file be loaded?
I'm trying to implement the DeepLab Model into a Tensorflow environment.
This is the github Repository: https://github.com/tensorflow/models/tree/master/research/deeplab
I'm using Python 3.6.8 and the Jupyter Notebook.
Now I'm facing the problem,…
Redshank
- 11
- 2
1
vote
0 answers
Deeplab v3 output changes when we change the input image size
I have custom trained deeplabv3 on set of images for detecting human contour. My crop size is 1025 x 513. I am able to train and export the trained model.
When I testing I get decent mask outputs when the input images are of size around 1025 x 450.…
Vikas
- 4,263
- 1
- 34
- 39
1
vote
2 answers
Deeplab - Inconsistent Inference vs Visualization Performance on trained Deeplab model
Describe the problem
I have successfully trained my model on a custom dataset with 4 classes of size 480x640, with an xception65 encoder, using Deeplab. I am getting decent results on the validation set whenever I use the vis.py script:…
Kayvon Khosrowpour
- 40
- 6
1
vote
1 answer
Semantic segmentation performance: Multi-class vs Two-class
We are developing an application for semantic segmentation for PC and Mobile Platform.Initially we tried the current state-of-the art DeeplabV3+ framework from google.The data-set used was Pascal VOC and the model from the official model zoo…
anilsathyan7
- 1,423
- 17
- 25