Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/gxd1994/TextBoxes-TensorFlow
TextBoxes re-implement using tensorflow
https://github.com/gxd1994/TextBoxes-TensorFlow
Last synced: 9 days ago
JSON representation
TextBoxes re-implement using tensorflow
- Host: GitHub
- URL: https://github.com/gxd1994/TextBoxes-TensorFlow
- Owner: gxd1994
- Created: 2017-03-17T11:33:15.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2017-03-17T04:53:21.000Z (over 7 years ago)
- Last Synced: 2024-08-02T11:16:04.208Z (3 months ago)
- Language: Python
- Size: 795 KB
- Stars: 43
- Watchers: 3
- Forks: 49
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# TextBoxes-TensorFlow
TextBoxes re-implementation using tensorflow.
This project is greatly inspired by [slim project](https://github.com/tensorflow/models/tree/master/slim)
And many functions are modified based on [SSD-tensorflow project](https://github.com/balancap/SSD-Tensorflow)
Later, we will overwrite this project so make it more
flexiable and modularized.Author:
Daitao Xing : [email protected]
Jin Huang : [email protected]# Progress
2017/ 03/14data_processing phase finished
Test:1. Download the dataset, put 1/ folder and gt.mat uner ddata/sythtext/ folder(will wirte script)
2. python datasets/data2record.py
3. python image_processing.py
output: batch_size * 300 * 300 * 3 image2017/ 03/17
Finish the design of training(can start training)
python train.py \
--train_dir=${TRAIN_DIR} \
--dataset_dir=${DATASET_DIR} \
--save_summaries_secs=60 \
--save_interval_secs=600 \
--weight_decay=0.0005 \
--optimizer=adam \
--learning_rate=0.001 \
--batch_size=32# Problems to be solved:
1. Need to redesign visualization
2. image_processing can be improved
# Next steps:
1. traing on other datasets
2. fine tunes
3. test
4. automatic downloading datasets and so on