参考：How to use setInitialMask on TrackerCSRT?
OpenCV, How to pass parameters into cv2.TrackerMedianFlow_create function?

opencv中kcf跟踪算子的使用方法

# 跟踪算子的初始化
tracker = cv2.TrackerKCF_create()
tracker.init(imgshow, (x, y, w, h))

# 跟踪
ret, bbox = tracker.update(imgshow)

如何自定义参数

import cv2

# 保存默认参数至制定文件
tracker = cv2.TrackerMedianFlow_create()
tracker.save('params.json')

# 从文件里面读取配置参数
tracker2 = cv2.TrackerMedianFlow_create()
fs = cv2.FileStorage("params.json", cv2.FileStorage_READ)
tracker2.read(fs.getFirstTopLevelNode())

KCF的参数配置

KCF的默认配置参数如下，但是具体各个参数负责什么，应该如何修改，还需要大家提供指导。

{
    "my_object": {
        "detect_thresh": 5.000000000000000e-01,
        "sigma": 2.0000000298023224e-01,
        "lambda": 9.9999997473787516e-05,
        "interp_factor": 7.5000002980232239e-02,
        "output_sigma_factor": 6.2500000000000000e-02,
        "resize": 1,
        "max_patch_size": 6400,
        "split_coeff": 1,
        "wrap_kernel": 0,
        "desc_npca": 1,
        "desc_pca": 2,
        "compress_feature": 1,
        "compressed_size": 2,
        "pca_learning_rate": 1.5000000596046448e-01
    }
}