Skip to content
A tensorflow implementation of object-contour-detection with fully convolutional encoder decoder network
Branch: master
Clone or download
Latest commit f73dede Oct 1, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.gitignore Initial commit Jun 4, 2018
000129.jpg Add files via upload Jun 4, 2018
000129.png Add files via upload Jun 4, 2018
000999.jpg Add files via upload Jun 4, 2018
000999.png Add files via upload Jun 4, 2018
LICENSE Initial commit Jun 4, 2018
README.md Update README.md Jun 4, 2018
create_labels.py Update create_labels.py Oct 1, 2018
eval.py Add files via upload Jun 4, 2018
model_contour.py Add files via upload Jun 4, 2018
ops.py Add files via upload Jun 4, 2018
requirements .txt Add files via upload Jun 4, 2018
train.py Add files via upload Jun 4, 2018
utils.py Add files via upload Jun 4, 2018

README.md

tensorflow-object-contour-detection

This is a tensorflow implimentation of Object Contour Detection with a Fully Convolutional Encoder-Decoder Network (https://arxiv.org/pdf/1603.04530.pdf) .

REQUIREMENTS :

pip install requirements.txt

Label Preparation :

To prepare the labels for contour detection from PASCAL Dataset , run create_lables.py and edit the file to add the path of the labels and new labels to be generated . Use this path for labels during training.

TRAINING :

python train.py \
    --max_to_keep=50 \
    --Epochs=100 \
    --momentum=0.9 \
    --learning_rate=.0000001 \
    --train_crop_size=480 \
    --clip_by_value=1.0 \
    --train_text = ${path to text file} \
    --log_dir = ${path to where logs will be saved} \
    --tf_initial_checkpoint=${PATH_TO_CHECKPOINT} \
    --label_dir = ${path to label directory} \
    --image_dir = ${path to image directory}

EVALUATION :

python eval.py \
    --checkpoint=${path to checkpoint to be evaluated} \
    --save_preds=${path to folder where predictions will be saved} \
    --image_dir = ${path to image directory} \
    --eval_crop_size=480 \
    --eval_text =  ${path to eval text file}

Results :

Image_1

prediction_1

Image_1

prediction_1

You can’t perform that action at this time.