PedestrianDetection

  • i5_313008
    了解作者
  • 1.6MB
    文件大小
  • zip
    文件格式
  • 0
    收藏次数
  • VIP专享
    资源类型
  • 0
    下载次数
  • 2022-04-12 07:48
    上传日期
介绍 在上的YOLOv3(Tensorflow后端)的。 指导 从下载YOLOv3权重。 将Darknet YOLO模型转换为Keras模型。 运行YOLO检测。 wget https://pjreddie.com/media/files/yolov3.weights python convert.py -w yolov3.cfg yolov3.weights model_data/yolo_weights.h5 python yolo.py OR python yolo_video.py [video_path] [output_path(optional)] 锚盒的选择 YOLO v3总共使用9个锚点框。 每个刻度三个。 如果在自定义数据集上训练YOLO,则必须使用K-Means聚类完成锚的生成。 然后,将锚按尺寸的降序排列。 为第一个音阶分配三个最大的锚点,为第二个音阶
PedestrianDetection-master.zip
  • PedestrianDetection-master
  • model_data
  • bdd100k_classes.txt
    7B
  • bdd100k_anchors.txt
    63B
  • font
  • FiraMono-Medium.otf
    124.4KB
  • SIL Open Font License.txt
    4.3KB
  • yolo3
  • __pycache__
  • utils.cpython-36.pyc
    3.8KB
  • model.cpython-36.pyc
    12.6KB
  • __init__.cpython-36.pyc
    167B
  • utils.py
    3.8KB
  • model.py
    16.1KB
  • __init__.cpython-36.pyc
    167B
  • train.py
    8.9KB
  • train.txt
    8.3MB
  • yolo.py
    8.3KB
  • LICENSE
    1KB
  • convert.py
    9.9KB
  • README.md
    1.6KB
  • kmeans.py
    3.4KB
  • yolov3.cfg
    8.1KB
  • bdd100k_annotation.py
    1.8KB
内容介绍
## Introduction A Keras implementation of YOLOv3 (Tensorflow backend) on [bdd100k dataset](http://bair.berkeley.edu/blog/2018/05/30/bdd/). --- ## Guide 1. Download YOLOv3 weights from [YOLO website](http://pjreddie.com/darknet/yolo/). 2. Convert the Darknet YOLO model to Keras model. 3. Run YOLO detection. ``` wget https://pjreddie.com/media/files/yolov3.weights python convert.py -w yolov3.cfg yolov3.weights model_data/yolo_weights.h5 python yolo.py OR python yolo_video.py [video_path] [output_path(optional)] ``` --- ## Choice of Anchor Boxes YOLO v3, in total uses 9 anchor boxes. Three for each scale. If training YOLO on a custom dataset, generation of anchors must be done using K-Means clustering. Then, arrange the anchors is descending order of dimensions. Assign the three biggest anchors for the first scale , the next three for the second scale, and the last three for the third. --- ## Training 1. Generate the annotation file using `python bdd100k_annotation.py` and class names file. One row for one image; Row format: `image_file_path box1 box2 ... boxN`; Box format: `x_min,y_min,x_max,y_max,class_id` (no space). For example: ``` path/to/img1.jpg 50,100,150,200,0 30,50,200,120,3 path/to/img2.jpg 120,300,250,600,2 ... ``` 2. Make sure you have run `python convert.py -w yolov3.cfg yolov3.weights model_data/yolo_weights.h5` The file model_data/yolo_weights.h5 is used to load pretrained weights. 3. Modify train.py and start training. `python train.py` Use your trained weights or checkpoint weights in yolo.py. Remember to modify class path or anchor path. ---
评论
    相关推荐