일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | ||||||
2 | 3 | 4 | 5 | 6 | 7 | 8 |
9 | 10 | 11 | 12 | 13 | 14 | 15 |
16 | 17 | 18 | 19 | 20 | 21 | 22 |
23 | 24 | 25 | 26 | 27 | 28 | 29 |
30 | 31 |
- V3
- 개발자
- 설치
- CycleGAN
- 욜로
- 리뷰
- ChatGPT
- 파이썬
- Detectron2
- 파워셀
- pyqt5
- 호흡분석
- NPY
- 언어모델
- 헬스케어
- 언리얼엔진
- ctypes
- 게임개발
- 딜러닝
- 딥러닝
- 텐서플로우
- python
- 논문
- 논문리뷰
- yolo
- TensorFlow
- connx
- C언어
- 프로그래머
- 이터널리턴
- Today
- Total
사냥꾼의 IT 노트
Pytorch를 이용한 YOLO v3 논문 구현 #6 - 최종 실행 본문
※본 포스팅은 아래 블로그를 참조해 번역하고 공부한 것입니다.
https://blog.paperspace.com/how-to-implement-a-yolo-object-detector-in-pytorch/
Tutorial on implementing YOLO v3 from scratch in PyTorch
Tutorial on building YOLO v3 detector from scratch detailing how to create the network architecture from a configuration file, load the weights and designing input/output pipelines.
blog.paperspace.com
YOLO v3를 구현하기 위한 모듈 및 라이브러리의 버전은 다음과 같습니다.
Requirements
- Windows 10
- python 3.10.6
- pytorch 1.12.1
- opencv 4.6.0
- cuda 11.3.1 및 cudnn
- six 1.16.0
Installation
GitHub - ayooshkathuria/YOLO_v3_tutorial_from_scratch: Accompanying code for Paperspace tutorial series "How to Implement YOLO v
Accompanying code for Paperspace tutorial series "How to Implement YOLO v3 Object Detector from Scratch" - GitHub - ayooshkathuria/YOLO_v3_tutorial_from_scratch: Accompanying code for Pap...
github.com
- yolo v3 weights 파일 다운로드: https://pjreddie.com/darknet/yolo/#:~:text=https%3A//pjreddie.com/media/files/yolov3.weights
YOLO: Real-Time Object Detection
YOLO: Real-Time Object Detection You only look once (YOLO) is a state-of-the-art, real-time object detection system. On a Pascal Titan X it processes images at 30 FPS and has a mAP of 57.9% on COCO test-dev. Comparison to Other Detectors YOLOv3 is extremel
pjreddie.com
Execution
- 본인이 설정한 가상환경 터미널을 열어줍니다. 저는 anaconda prompt를 사용했습니다.
- 가상환경에 설치된 라이브러리와 모듈 버전을 확인합니다.
- 깃허브에서 다운로드한 파일들을 확인합니다. 가중치 파일도 같은 폴더에 넣어줍니다.
- 이미지 확인: 코드 입력으로 객체 검출을 시행할 수 있습니다.
python detect.py --images imgs/woman-car.png --det det/
정상 실행됐을 경우의 프롬프트 화면입니다.
- predicted in 2.168 seconds: 예측하는데 2.168초 걸림
- Objects Detected peson car: coco datasets의 클래스 중 하나인 peson과 car 객체가 검출 됨
- Task: 각 로직 실행에 걸린 시간 표시
det 디렉토리 밑에 output 이미지가 det_을 붙인채로 저장된 걸 확인할 수 있습니다.
- 비디오 확인: 코드 입력으로 객체 검출을 시행할 수 있습니다.
python video.py --video video/test.mp4
https://www.youtube.com/watch?v=zHTK7GJ6D8A
비디오 객체 검출을 녹화해 유튜브에 올렸습니다. 객체가 잘 검출되는 걸 확인할 수 있습니다.
'YOLO' 카테고리의 다른 글
Human Respiratory Signals based on 1D YOLO Model (0) | 2024.06.06 |
---|---|
Pytorch를 이용한 YOLO v3 논문 구현 #5 - input, output (0) | 2022.09.27 |
Pytorch를 이용한 YOLO v3 논문 구현 #4 - NMS 구현과 objectness score (0) | 2022.09.27 |
Pytorch를 이용한 YOLO v3 논문 구현 #3 - weights file 이해하기 (0) | 2022.09.27 |
Pytorch를 이용한 YOLO v3 논문 구현 #2 - 신경망 순전파 구현 (0) | 2022.09.27 |