CV-12 TEAM YOLO
κΉμΉν_T3038 | λ Έμ°½ν_T3074 | μ΅μ©μ_T3219 | μ΅μ§μ_T3225 | μ΅νλ‘_T3228 |
---|---|---|---|---|
![]() |
![]() |
![]() |
![]() |
![]() |
- μ¬κ³ νμ μλν λ° νλ‘μΈμ€ ν¨μ¨νλ₯Ό μν Computer Vision μμ€ν
- λͺ¨λ 맀μ₯μμ 맀λ νλͺ©μ κ΄λ¦¬νλ μμ€ν μΌλ‘ μ¬μ©ν μ μμ
Python Video_test.py --video_name {TestVideo κ²½λ‘} --output_name {Output Video name} --topdown {bool}
Train / Validation Image
Test
Task | Model | Performance |
---|---|---|
Object detection | YOLOv5 | mAP50: 0.9915 |
Classification | ResNet18 | accuracy: 0.966 |
κ²°νκ°μ§ | YOLO+RESNET18+SSIM | mAP50: 0.45 |
|-- README.md
|-- Video_test.py
|-- backend
| |-- Send_To_Video.py
| |-- backend_flask_detector.py
|-- dataset
| |-- README.md
| |-- images
| |-- instances_Test.json
| |-- instances_Train.json
| |-- instances_Validation.json
|-- evaluation
| |-- mAP.ipynb
|-- front
| |-- front_table.py
| |-- result.csv
|-- modules
| |-- Human_detect.py
| |-- calcIou.py
| |-- changeDetector.py
| |-- classification.py
| |-- model.py
| |-- outDetection.py
| |-- ramen_detect.py
| |-- rgb_simularity.py
| |-- utils.py
|-- requirements.txt
|-- train
| |-- classification
| |-- yolov5
|-- weights
| |-- classification_all.pth
| |-- outDetector.pth
| |-- ramen_best.pt
yolov5==6.1.2
streamlit==1.9.0
streamlit-webrtc==0.37.0
torchvision==0.8.1
scikit-image==0.19.2
albumentations==1.1.0
flask==2.1.2
torch==1.11.0
uvicorn==0.17.6