Awesome Open Source
Awesome Open Source

https://github.com/tensorflow/tensorflow

https://github.com/ildoonet/tf-pose-estimation

Denis Tome, Chris Russell, Lourdes Agapito提出的Convolutional 3D Pose Estimation from a Single Image论文

``````[zyxcambridge](https://github.com/zyxcambridge) 童鞋
[manoshape](https://github.com/manoshape) 童鞋
[seriouslyhao](https://github.com/seriouslyhao) 童鞋
本项目模型由zyxcambridge、manoshape、seriouslyhao提供
``````

TensorFlowObjectDetectionAPIModel 为检测规则画框

TensorFlowImageClassifier2 为车道检测之后不规则绘制（因时间仓促 还没有进行绘图优化） 识别道路的测试方法请自行百度寻找图片或者视频都可以

TensorFlowImageClassifier3 是用来识别人体骨架的 这个模型是有特定输入和特定输出的 需要经过3层转换 才能使用 接下来准备上线道路障碍物识别... 最新版骨架识别目前支持区分各个身体部位具体情况请看注释

Camera2BasicFragment4 这是一个用检测来识别车道和前车 里面增加了点逻辑来判断是否是车道偏离或者前车过近 具体做法是 如果检测出线则判断斜率k = (y2-y1)/(x2-x1)然后设定一个固定斜率来判断是否是车道偏离 如果是检测出前面的车辆中心点在横屏8分之2到8分之6的范围内则判断中心点居上距离大于一定范围则算前车过近 或者如果车的高度大于一定级别则算前车过近

WH19每一层的解释

``````         Nose = 0
Neck = 1
RShoulder = 2
RElbow = 3
RWrist = 4
LShoulder = 5
LElbow = 6
LWrist = 7
RHip = 8
RKnee = 9
RAnkle = 10
LHip = 11
LKnee = 12
LAnkle = 13
REye = 14
LEye = 15
REar = 16
LEar = 17
Background = 18
``````

``````    CocoPairs = [
(1, 2), (1, 5), (2, 3), (3, 4), (5, 6), (6, 7), (1, 8), (8, 9), (9, 10), (1, 11),
(11, 12), (12, 13), (1, 0), (0, 14), (14, 16), (0, 15), (15, 17), (2, 16), (5, 17)
]   # = 19
``````

``````        CocoPairsNetwork = [
(12, 13), (20, 21), (14, 15), (16, 17), (22, 23), (24, 25), (0, 1), (2, 3), (4, 5),
(6, 7), (8, 9), (10, 11), (28, 29), (30, 31), (34, 35), (32, 33), (36, 37), (18, 19), (26, 27)
]  # = 19
``````

``````    求出Dx = x1 - x2
Dy = y1 - y2
``````

``````    k = 根号下（Dx平方-Dy平方）
``````

``````    for(i = x1; i <= x2; i += dx/10)
x[n] = i;
Y同理
``````

``````    分数>0.2的个数乘 >0.2的分数相加 得到最终分数，并且保留有多少个>0.2的
之前两点间距离小于0.0001分数为0的个数也为0
``````

Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
java (30,097
android (6,011
tensorflow (2,071
artificial-intelligence (575
demo (474
ai (438
app (367
detection (219
phone (76
recognition (56