Level2 P-stage Object Detection

💡 Team: 컴퓨터구조

Project Overview

Predict Trash Objects
Input: 1024 x 1024 Image
Output: Object annotations
- Class(10): General trash, Paper, Paper pack, Metal, Glass, Plastic, Styrofoam, Plastic bag, Battery, Clothing
- Confidence score: 0 ~ 1
- Object coordinate: (xmin, ymin, xmax, ymax)

Archive contents

image-classification-level1-02/
├── dataset/
│   ├── train/ (4883 images)
│   ├── test/  (4871 images)
│   ├── train.json
│   └── test.json
├── mmdetection/
│   ├── configs/
│   ├── mmdetection library files
│   ├── train.py
│   └── inference.py
└── util/

dataset/ : download from https://stages.ai/

get start

train & inference

cd mmdetection

python train.py
python inference.py

visualize

cd util

jupyter notebook Visualize.ipynb

set result csv in second shell

it also has Visualize_val_gt -> Visualize ground truth of train data & valid

it also has Visualize_val -> Visualize result of valid

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
김성민 @ksm0517
박지민 @박지민
박진형 @ppjh8263
심세령 @Seryoung Shim
윤하정 @Yoon Hajung

Data Citation `네이버 커넥트재단 - 재활용 쓰레기 데이터셋 / CC BY 2.0`

About

object-detection-level2-cv-07 created by GitHub Classroom

Languages

Language:Jupyter Notebook 78.7%Language:Python 20.5%Language:Cuda 0.3%Language:C++ 0.2%Language:Shell 0.2%Language:Dockerfile 0.0%Language:JavaScript 0.0%Language:Makefile 0.0%Language:Batchfile 0.0%Language:CMake 0.0%