How can labeling my own dataset

kerolos commented 6 years ago

How can I label a customized dataset to be able to use in Caffe_segnet? Which tool or program I should use to create my own label dataset with the same format as dataset used in this tutorial: 'http://mi.eng.cam.ac.uk/projects/segnet/tutorial.html'. I found that it is Gray images had been created to train the model (trainannot), furthermore, the Softmax layer is created by class_weighting parameters for each class, which should represent the label in this folder trainannot. My question is, how can i label my own dataset to train the bixel-wise classification ?

softmax layer in this file 'segnet_train.prototxt' layer { name: "loss" type: "SoftmaxWithLoss" bottom: "conv1_1_D" bottom: "label" top: "loss" softmax_param {engine: CAFFE} loss_param: { weight_by_label_freqs: true ignore_label: 11 class_weighting: 0.2595 class_weighting: 0.1826 class_weighting: 4.5640 class_weighting: 0.1417 class_weighting: 0.9051 class_weighting: 0.3826 class_weighting: 9.6446 class_weighting: 1.8418 class_weighting: 0.6823 class_weighting: 6.2478 class_weighting: 7.3614 } }

xhy3054 commented 6 years ago

did you solve it?

ArunJ1 commented 5 years ago

I have same issue

Somebody pls help

ArunJ1 commented 5 years ago

I also face same problem...Any help??? by Arun

ArunJ1 commented 5 years ago

I have same issue

Somebody pls help

alexgkendall / caffe-segnet

How can labeling my own dataset #129