- Jeongtae Shin(@Klassikcat): Project Management, Model build, Engineering
- Jaewoong Lee(@colin19950703): Augmentation, model build, Enigeering
- Hongkyun Kim(@ghdrbs0302): Data Inspection, Test data Labeling, Engineering
!pip install -U git+https://github.com/Klassikcat/project-NEXTLab-CNN-EfficientNet
https://drive.google.com/file/d/1UyVnR5Pi7NvzD3LPnkllGEyn8afMi5jF/view?usp=sharing
project-NEXTLab
┖ core
┖ config.py # Tensorflow Configuration
┖ utils.py
┖ model
┖ EfficientNet.py # Model block and builder
┖ initializers.py # from EfficientNetv2 (https://arxiv.org/abs/2104.00298)
┖ layer.py # from EfficietnNetv2
┖ model.py # from EfficientNetv2
┖ params.py # from EfficientNetv2
┖ utils
┖ augmentation # Data augmentation tools
┖ augmentation.py
┖ launcher_augment.py
┖ launcher_classname.py
┖ from_folder
┖ loader.py # get train/valid/test dataset from directory
┖ tfrecord
┖ tfrecordMaker # Make TFRecord
┖ tfrecordViewer # Visualize, Load TFRecord
┖ unicodeUtil.py # Korean utilities assemble jamo
┖ visuialization.py # Plots(data distribution, heatmap)
setup.py # installer
readme.md
- Making EfficientNetv2 application that distinguish 322 cars by using photos captured by traffic camera
- Traffic enforcement, Parking lot management.
- Uses modified "자동차 차종/연식/번호판 인식용 영상" from AIHub: https://aihub.or.kr/aidata/27727
- Uses 522 classes in "세단", "SUB", "해치백", "승합"
- Road Camera dataset from NEXTLab: https://www.nextlab.ai/
- 5016 images belonging to 80 classes
approximately 82% of data is korean cars(hyundai, KIA)
approximately 75% of data is korean cars(sonata, etc...)
51% sedan, 28.8% SUV, 10% Hatchback, 9% van
mainly consists with 2010~2018 cars
- use these augmentations taking the possible road situations into account.
- Drop the class with less than 30 pieces of data in order to avoid data leakage
- randomly pick 300 data. if the class has less than 30 data, proceed Data Augmentation for equal distribution of data
- EfficientNetB0(224, 224) with 30 layer freeze
- train: 0.9051/0.2755
- valid: 0.8960/0.3769
- train: 0.9880
- valid: 0.9717
[ ] Increase valid/test accuracy by making Ensemble model
[ ] Label Smoothing
[ ] Fix all false label