Comments (6)
For deeplab pre-trained models, I believe the order of image channels is BGR.
I provide the pre-trained resnet models from TF Slim, where the means should not be divided by 255.
https://github.com/tensorflow/models/blob/master/research/slim/preprocessing/vgg_preprocessing.py
And you are right about the order. If resnet pre-trained models are used, the order should be changed back to RGB. A one-line change is enough.
from deeplab-v2--resnet-101--tensorflow.
But actually I don't think it is crucial. In this task, the size of training patches is also different from that in resnet. And the set of images is different. Maybe simply using image_mean=[127.5, 127.5, 127.5] will work well.
from deeplab-v2--resnet-101--tensorflow.
Agree. I have tested with difference IMAGE_MEAN and it has no much performance diffference. Could I ask ome more question about pretrained model? When you use resnet pretrained model, it means that you will copy pretrained weight of resnet to encoder part of deeplab network. Then we will train the decoder part. But I found that you also trained the encoder part after copy weight from resnet model. Why did not only train the decoder part? Thanks
from deeplab-v2--resnet-101--tensorflow.
This is because the set of images changes. It is true that they are all natural images with similar features so that transfer learning is feasible here. However, images in PASCAL or CITYSCAPES do not appear in ImageNet. Thus, we'd like to fine-tune the encoder to let it fit the new set of images. Actually, we use the pre-trained models in order to make sure the training converge, as the number of images in PASCAL or CITYSCAPES is much smaller than that in ImageNet.
from deeplab-v2--resnet-101--tensorflow.
I see. So it looks like we use pretrain model to have a good weight initialization. Then retrain the model with the good weight. Am I right?
from deeplab-v2--resnet-101--tensorflow.
Yes.
from deeplab-v2--resnet-101--tensorflow.
Related Issues (20)
- Optimizer choice: Adam VS SGD HOT 1
- Another dataset HOT 4
- Problem with pre-trained model HOT 1
- Results for VOC2012 are not correct HOT 3
- Training Cityscapes - Changes in label_utils.py HOT 1
- pretrain model download
- How to process the outline of the object in Segmentation label images? HOT 1
- The problem about paper "Smoothed Dilated Convolutions for Improved Dense Prediction". HOT 1
- Which part of your code corresponds to the CRF in Deeplab V2? HOT 1
- How to predict dynamically from graph HOT 1
- A question about paper "Smoothed Dilated Convolutions for Improved Dense Prediction" HOT 2
- NotFoundError (see above for traceback): Tensor name HOT 3
- only can train 2images HOT 1
- Can I input 6 channel images for training? HOT 2
- how about results(mIoU) on validation set and test set HOT 2
- Hi, could you please implement it on cityscape? HOT 3
- how about the hyperparms of cityscape
- ABOUT BATCHSIZE WHEN TRAINING ON CITYSCAPES HOT 1
- It is OOM. Try reducing input_height and input_width. HOT 1
- resnet101 ------- miou 70.7% HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from deeplab-v2--resnet-101--tensorflow.