/semantic-segmentation

semantic-segmentation-level2-cv-07 created by GitHub Classroom

Primary LanguagePython

Level2 P-stage Semantic Segmentation

πŸ’‘ Team: 컴퓨터ꡬ쑰

Project Overview

  • Trash semantic segmentation
  • Input: 512 x 512 Image
    • Train: 2617 images
    • Validation: 655 images
  • Output: Category classification for each pixel
    • Class(11): Background, General trash, Paper, Paper pack, Metal, Glass, Plastic, Styrofoam, Plastic bag, Battery, Clothing

Archive contents

image-classification-level1-02/
β”œβ”€β”€ input/data/
β”‚   β”œβ”€β”€ batch_01_vt/
β”‚   β”œβ”€β”€ batch_02_vt/
β”‚   β”œβ”€β”€ batch_03/
β”‚   β”œβ”€β”€ train.json
β”‚   β”œβ”€β”€ val.json
β”‚   β”œβ”€β”€ train_all.json
β”‚   └── test.json
β”œβ”€β”€ mmsegmentation/
β”‚   β”œβ”€β”€ configs/
β”‚   β”œβ”€β”€ mmdetection library folders and files ...
β”‚   β”œβ”€β”€ train.py
β”‚   └── inference.py
└── util/

get start

train & inference

cd mmsegmentation

python train.py
python inference.py

visualize

cd util

converting visualize_CSVs py to ipynb

jupyter notebook visualize_CSVs.ipynb

# set result csv files
csv_names = ['Unet_resnet50.csv', 'deeplabv3_resnet50_31epoch.csv']

Requirements

  • Ubuntu 18.04.5
  • Python 3.8.5
  • pytorch 1.7.1
  • torchvision 0.8.2

Install packages : pip install -r requirements.txt

Hardware

  • CPU: 8 x Intel(R) Xeon(R) Gold 5220 CPU
  • GPU: V100
  • RAM: 88GB

Contributors

Name @github
고재욱 @고재욱
κΉ€μ„±λ―Ό @ksm0517
박지민 @박지민
λ°•μ§„ν˜• @ppjh8263
심세령 @Seryoung Shim
μœ€ν•˜μ • @Yoon Hajung

Data Citation
넀이버 컀λ„₯νŠΈμž¬λ‹¨ - μž¬ν™œμš© μ“°λ ˆκΈ° 데이터셋 / CC BY 2.0