객체 탐지 : Yolov5

• Yolov5

  You only look once, Localization과 Classification이 동시에 진행되는 one-stage detection 모델. 탐지 속도가 빠르면서도 정확도가 높은 장점이 있다. 여러 버전이 존재하고, 이용하는 v5는 pytorch로 구현되어있는 대표적인 오픈소스

• License

  GPL-3.0 license

Model

사용 모델 선택

<aside> 🌟 Recommendation on YoloV5 Tutorials For mobile deployments we recommend YOLOv5s/m, for cloud deployments we recommend YOLOv5l/x.

</aside>

from n params module arguments 0 -1 1 3520 models.common.Conv [3, 32, 6, 2, 2] 1 -1 1 18560 models.common.Conv [32, 64, 3, 2] 2 -1 1 18816 models.common.C3 [64, 64, 1] 3 -1 1 73984 models.common.Conv [64, 128, 3, 2] 4 -1 2 115712 models.common.C3 [128, 128, 2] 5 -1 1 295424 models.common.Conv [128, 256, 3, 2] 6 -1 3 625152 models.common.C3 [256, 256, 3] 7 -1 1 1180672 models.common.Conv [256, 512, 3, 2] 8 -1 1 1182720 models.common.C3 [512, 512, 1] 9 -1 1 656896 models.common.SPPF [512, 512, 5] 10 -1 1 131584 models.common.Conv [512, 256, 1, 1] 11 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest'] 12 [-1, 6] 1 0 models.common.Concat [1] 13 -1 1 361984 models.common.C3 [512, 256, 1, False] 14 -1 1 33024 models.common.Conv [256, 128, 1, 1] 15 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest'] 16 [-1, 4] 1 0 models.common.Concat [1] 17 -1 1 90880 models.common.C3 [256, 128, 1, False] 18 -1 1 147712 models.common.Conv [128, 128, 3, 2] 19 [-1, 14] 1 0 models.common.Concat [1] 20 -1 1 296448 models.common.C3 [256, 256, 1, False] 21 -1 1 590336 models.common.Conv [256, 256, 3, 2] 22 [-1, 10] 1 0 models.common.Concat [1] 23 -1 1 1182720 models.common.C3 [512, 512, 1, False] 24 [17, 20, 23] 1 229245 models.yolo.Detect [80, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]] Model summary: 214 layers, 7235389 parameters, 7235389 gradients

Dataset

<aside> 🌟 Recommendation on YoloV5 Tutorials

• Images per class. ≥ 1500 images per class recommended
• Instances per class. ≥ 10000 instances (labeled objects) per class recommended
• Label consistency. All instances of all classes in all images must be labelled. Partial labelling will not work. (이미지에 있는 모든 클래스들이 라벨링 되어야함)
• Background images. Background images are ****images with no objects that are added to a dataset to reduce False Positives (FP). We recommend about 0-10% background images to help reduce FPs. No labels are required for background images. (해당 이미지에 객체가 없는 데이터가 10%이내로 필요함, roboflow에서 marked as null) </aside>

Training

<aside> 🌟 Recommendation on YoloV5 Tutorials

• Epochs. Start with 300 epochs. 그리고 오버피팅되면 줄이고 아니면 늘리기
• Image size. COCO trains at native resolution of -img 640 Best inference results are obtained at the same -img as the training was run at, i.e. if you train at -img 1280 you should also test and detect at -img 1280.
• Batch size. Use the largest -batch-size that your hardware allows for. </aside>

Hyperparameter Evolution

hyp.yaml

• evolve + data 검수 결과