Init the repo

.gitignore

@@ -0,0 +1,2 @@
+
+__pycache__/

README.md

@@ -1,2 +1,25 @@
 # DeepHoughTransform
 Add training codes into Hanqer/deep-hough-transform.
+
+### Brief 
+- Since there are no training codes in the original repo, I decided to add some training codes.
+- I considered to use the architecture of deep-hough-transform to train on my own task. The task is not actually similar to the task of `Semantic Line Detection`.
+- This repo is just an attempt of writing the training codes. No plan to open my dataset or evaluate the model.
+- _(In Chinese):_ 这条用中文再啰嗦叙述一下，因为我个人目前的工作想要考虑看一下这个deep hough transform的方法能不能用在我的数据集上，但是他这个没有训练代码啊，自己试着写一下，不保证准确率那些。
+- 多说一句，看代码的时候我发现trainset的设置（数据读取、transform）作者闭源了，另外label和loss计算也都闭源了，所以我考虑这两个点是有比较重点的tricks。
+
+### Requirements
+- Developed under `Pytorch1.3 Python3.6`
+
+
+### Remarks
+- (这段用中文记录一下复写的记录，后面可能会整理出一篇博客把作者的工作分析一下，另外补充一下代码实现细节)
+- 截止目前为止(9/11/2020)，训练代码没有开源，除了网络结构之外，能够分析一下前后处理的就是forward.py脚本
+```
+key_points = model(images)
+```
+说明模型前向输出的是这个keypoints
+
+### TODOs
+- Not finished yet
+

READMEori.md

@@ -0,0 +1,39 @@
+### Difference (Written by Una 07/31/2020)
+- Add train codes (uses CrossEntropy according to the paper)
+- "AverageMeter" refer to [InnovArul/pytorch_utils](https://github.com/InnovArul/pytorch_utils/blob/c58115a88f4ad90bdb2f69f85998cdc0fa43831b/generic/avgmeter.py)
+
+---
+Code for paper "Deep Hough Transform for Semantic Line Detection" (ECCV2020).
+### Deep Hough Transform
+![pipeline](./pipeline.png)
+### Requirements
+``` 
+numpy
+scipy
+opencv-python
+scikit-image
+pytorch>=1.0
+torchvision
+tqdm
+yml
+deep-hough
+```
+
+To install deep-hough, run the following commands.
+```sh
+cd deep-hough-transform
+cd model/_cdht
+python setup.py build 
+python setup.py install --user
+```
+Pretrain model (based on ResNet50-FPN): https://drive.google.com/file/d/1a6Rbu1Bslyo9sjNlUUdi7NnSTdRIWwS5/view?usp=sharing
+### Forward
+Generate visualization results and save coordinates to _.npy file.
+```sh
+CUDA_VISIBLE_DEVICES=0 python forward.py --model （your_best_model.pth） --tmp (your_result_save_dir)
+```
+### Test
+Test the EA-score on SEL dataset. After forwarding the model and get the coordinates files. Run the following command to produce EA-score.
+```sh
+python test.py --pred result/debug/visualize_test/(change to your onw path which includes _.npy files) --gt gt_path/include_txt
+```

basic_ops.py

@@ -0,0 +1,85 @@
+import cv2
+import numpy as np
+
+class Line(object):
+    def __init__(self, coordinates=[0, 0, 1, 1]):
+        """
+        coordinates: [y0, x0, y1, x1]
+        """
+        assert isinstance(coordinates, list)
+        assert len(coordinates) == 4
+        assert coordinates[0]!=coordinates[2] or coordinates[1]!=coordinates[3]
+        self.__coordinates = coordinates
+
+    @property
+    def coord(self):
+        return self.__coordinates
+
+    @property
+    def length(self):
+        start = np.array(self.coord[:2])
+        end = np.array(self.coord[2::])
+        return np.sqrt(((start - end) ** 2).sum())
+
+    def angle(self):
+        y0, x0, y1, x1 = self.coord
+        if x0 == x1:
+            return -np.pi / 2
+        return np.arctan((y0-y1) / (x0-x1))
+
+    def rescale(self, rh, rw):
+        coor = np.array(self.__coordinates)
+        r = np.array([rh, rw, rh, rw])
+        self.__coordinates = np.round(coor * r).astype(np.int).tolist()
+
+    def __repr__(self):
+        return str(self.coord)
+
+
+def get_boundary_point(y, x, angle, H, W):
+    '''
+    Given point y,x with angle, return a two point in image boundary with shape [H, W]
+    return point:[x, y]
+    '''
+    point1 = None
+    point2 = None
+
+    if angle == -np.pi / 2:
+        point1 = (x, 0)
+        point2 = (x, H-1)
+    elif angle == 0.0:
+        point1 = (0, y)
+        point2 = (W-1, y)
+    else:
+        k = np.tan(angle)
+        if y-k*x >=0 and y-k*x < H:  #left
+            if point1 == None:
+                point1 = (0, int(y-k*x))
+            elif point2 == None:
+                point2 = (0, int(y-k*x))
+                if point2 == point1: point2 = None
+        # print(point1, point2)
+        if k*(W-1)+y-k*x >= 0 and k*(W-1)+y-k*x < H: #right
+            if point1 == None:
+                point1 = (W-1, int(k*(W-1)+y-k*x))
+            elif point2 == None:
+                point2 = (W-1, int(k*(W-1)+y-k*x)) 
+                if point2 == point1: point2 = None
+        # print(point1, point2)
+        if x-y/k >= 0 and x-y/k < W: #top
+            if point1 == None:
+                point1 = (int(x-y/k), 0)
+            elif point2 == None:
+                point2 = (int(x-y/k), 0)
+                if point2 == point1: point2 = None
+        # print(point1, point2)
+        if x-y/k+(H-1)/k >= 0 and x-y/k+(H-1)/k < W: #bottom
+            if point1 == None:
+                point1 = (int(x-y/k+(H-1)/k), H-1)
+            elif point2 == None:
+                point2 = (int(x-y/k+(H-1)/k), H-1)
+                if point2 == point1: point2 = None
+        # print(int(x-y/k+(H-1)/k), H-1)
+        if point2 == None : point2 = point1
+    return point1, point2
+

data/SEL_dataset/test_gt/0015.txt

@@ -0,0 +1,2 @@
+1.0, 203.4, 400.0, 285.0, 400.0, 400.0
+1.0, 179.0, 400.0, 190.0, 400.0, 400.0

data/SEL_dataset/test_gt/0015new.txt

@@ -0,0 +1,2 @@
+1.0, 203.4, 400.0, 285.0, 400.0, 400.0
+1.0, 179.0, 400.0, 190.0, 400.0, 400.0

data/SEL_dataset/test_gt/0033.txt

@@ -0,0 +1 @@
+1.0, 270.9, 400.0, 148.1, 400.0, 400.0

data/SEL_dataset/test_gt/0033new.txt

@@ -0,0 +1 @@
+1.0, 270.9, 400.0, 148.1, 400.0, 400.0

data/SEL_dataset/test_gt/0046.txt

@@ -0,0 +1,3 @@
+1.0, 166.9, 400.0, 184.7, 400.0, 400.0
+1.0, 197.0, 400.0, 197.0, 400.0, 400.0
+1.0, 227.2, 400.0, 237.9, 400.0, 400.0

data/SEL_dataset/test_gt/0046new.txt

@@ -0,0 +1,3 @@
+1.0, 166.9, 400.0, 184.7, 400.0, 400.0
+1.0, 197.0, 400.0, 197.0, 400.0, 400.0
+1.0, 227.2, 400.0, 237.9, 400.0, 400.0

data/SEL_dataset/test_gt/0058.txt

@@ -0,0 +1,4 @@
+1.0, 145.9, 400.0, 92.6, 400.0, 400.0
+1.0, 161.8, 400.0, 188.2, 400.0, 400.0
+1.0, 94.5, 400.0, 66.9, 400.0, 400.0
+1.0, 239.0, 400.0, 197.9, 400.0, 400.0

data/SEL_dataset/test_gt/0058new.txt

@@ -0,0 +1,4 @@
+1.0, 145.9, 400.0, 92.6, 400.0, 400.0
+1.0, 161.8, 400.0, 188.2, 400.0, 400.0
+1.0, 94.5, 400.0, 66.9, 400.0, 400.0
+1.0, 239.0, 400.0, 197.9, 400.0, 400.0

data/SEL_dataset/test_gt/0092.txt

@@ -0,0 +1 @@
+1.0, 307.7, 400.0, 27.8, 400.0, 400.0

data/SEL_dataset/test_gt/0092new.txt

@@ -0,0 +1 @@
+1.0, 307.7, 400.0, 27.8, 400.0, 400.0

data/SEL_dataset/test_gt/0095.txt

@@ -0,0 +1 @@
+1.0, 93.7, 400.0, 270.0, 400.0, 400.0

data/SEL_dataset/test_gt/0095new.txt

@@ -0,0 +1 @@
+1.0, 93.7, 400.0, 270.0, 400.0, 400.0

data/SEL_dataset/test_gt/0107.txt

@@ -0,0 +1 @@
+1.0, 145.0, 400.0, 145.0, 400.0, 400.0

data/SEL_dataset/test_gt/0107new.txt

@@ -0,0 +1 @@
+1.0, 145.0, 400.0, 145.0, 400.0, 400.0

data/SEL_dataset/test_gt/0111.txt

@@ -0,0 +1 @@
+1.0, 183.5, 400.0, 114.5, 400.0, 400.0

data/SEL_dataset/test_gt/0111new.txt

@@ -0,0 +1 @@
+1.0, 183.5, 400.0, 114.5, 400.0, 400.0

data/SEL_dataset/test_gt/0114.txt

@@ -0,0 +1,3 @@
+1.0, 193.7, 400.0, 198.1, 400.0, 400.0
+1.0, 244.4, 400.0, 257.2, 400.0, 400.0
+1.0, 330.4, 400.0, 366.7, 400.0, 400.0

data/SEL_dataset/test_gt/0114new.txt

@@ -0,0 +1,3 @@
+1.0, 193.7, 400.0, 198.1, 400.0, 400.0
+1.0, 244.4, 400.0, 257.2, 400.0, 400.0
+1.0, 330.4, 400.0, 366.7, 400.0, 400.0

data/SEL_dataset/test_gt/0150.txt

@@ -0,0 +1,2 @@
+1.0, 165.7, 400.0, 297.6, 400.0, 400.0
+1.0, 379.0, 400.0, 374.9, 400.0, 400.0

data/SEL_dataset/test_gt/0153.txt

@@ -0,0 +1,2 @@
+1.0, 253.9, 400.0, 258.0, 400.0, 400.0
+1.0, 295.8, 400.0, 323.3, 400.0, 400.0

data/SEL_dataset/test_gt/0198.txt

@@ -0,0 +1 @@
+1.0, 339.0, 400.0, 71.7, 400.0, 400.0

data/SEL_dataset/test_gt/0199.txt

@@ -0,0 +1,2 @@
+1.0, 208.1, 400.0, 203.0, 400.0, 400.0
+1.0, 146.1, 400.0, 38.5, 400.0, 400.0

data/SEL_dataset/test_gt/0215.txt

@@ -0,0 +1,2 @@
+1.0, 266.4, 400.0, 244.0, 400.0, 400.0
+1.0, 340.4, 400.0, 316.7, 400.0, 400.0

data/SEL_dataset/test_gt/0220.txt

@@ -0,0 +1 @@
+1.0, 124.5, 400.0, 116.8, 400.0, 400.0

data/SEL_dataset/test_gt/0239.txt

@@ -0,0 +1,2 @@
+1.0, 294.0, 393.7, 1.0, 400.0, 400.0
+1.0, 310.2, 400.0, 68.3, 400.0, 400.0

data/SEL_dataset/test_gt/0251.txt

@@ -0,0 +1 @@
+1.0, 116.0, 400.0, 116.0, 400.0, 400.0

data/SEL_dataset/test_gt/0283.txt

@@ -0,0 +1 @@
+1.0, 206.4, 400.0, 218.3, 400.0, 400.0

data/SEL_dataset/test_gt/0286.txt

@@ -0,0 +1,2 @@
+1.0, 295.8, 400.0, 20.6, 400.0, 400.0
+1.0, 208.2, 400.0, 84.9, 400.0, 400.0

data/SEL_dataset/test_gt/0311.txt

@@ -0,0 +1 @@
+1.0, 173.0, 400.0, 173.0, 400.0, 400.0

data/SEL_dataset/test_gt/0315.txt

@@ -0,0 +1 @@
+1.0, 305.0, 400.0, 200.1, 400.0, 400.0

data/SEL_dataset/test_gt/0319.txt

@@ -0,0 +1,3 @@
+1.0, 257.5, 400.0, 228.9, 400.0, 400.0
+1.0, 28.5, 400.0, 126.8, 400.0, 400.0
+1.0, 268.5, 400.0, 289.9, 400.0, 400.0

data/SEL_dataset/test_gt/0333.txt

@@ -0,0 +1 @@
+1.0, 256.9, 400.0, 260.3, 400.0, 400.0

data/SEL_dataset/test_gt/0342.txt

@@ -0,0 +1,2 @@
+1.0, 134.9, 400.0, 141.0, 400.0, 400.0
+1.0, 128.5, 400.0, 98.9, 400.0, 400.0

data/SEL_dataset/test_gt/0348.txt

@@ -0,0 +1 @@
+1.0, 180.8, 400.0, 211.5, 400.0, 400.0

data/SEL_dataset/test_gt/0351.txt

@@ -0,0 +1,2 @@
+1.0, 111.9, 400.0, 137.5, 400.0, 400.0
+1.0, 309.9, 400.0, 287.7, 400.0, 400.0

data/SEL_dataset/test_gt/0353.txt

@@ -0,0 +1,2 @@
+1.0, 274.9, 400.0, 248.0, 400.0, 400.0
+1.0, 305.0, 400.0, 305.0, 400.0, 400.0

data/SEL_dataset/test_gt/0357.txt

@@ -0,0 +1,2 @@
+1.0, 232.7, 400.0, 183.3, 400.0, 400.0
+1.0, 263.9, 400.0, 235.4, 400.0, 400.0

data/SEL_dataset/test_gt/0378.txt

@@ -0,0 +1 @@
+1.0, 309.2, 295.2, 1.0, 400.0, 400.0

data/SEL_dataset/test_gt/0393.txt

@@ -0,0 +1 @@
+1.0, 287.4, 400.0, 282.7, 400.0, 400.0

data/SEL_dataset/test_gt/0399.txt

@@ -0,0 +1,3 @@
+1.0, 156.1, 400.0, 112.8, 400.0, 400.0
+1.0, 207.3, 400.0, 233.3, 400.0, 400.0
+1.0, 286.4, 400.0, 322.0, 400.0, 400.0

data/SEL_dataset/test_gt/0400.txt

@@ -0,0 +1,2 @@
+1.0, 285.4, 400.0, 261.3, 400.0, 400.0
+1.0, 381.7, 400.0, 359.1, 400.0, 400.0

data/SEL_dataset/test_gt/0416.txt

@@ -0,0 +1 @@
+1.0, 239.5, 400.0, 193.3, 400.0, 400.0

data/SEL_dataset/test_gt/0432.txt

@@ -0,0 +1 @@
+1.0, 231.5, 400.0, 217.0, 400.0, 400.0

data/SEL_dataset/test_gt/0435.txt

@@ -0,0 +1,3 @@
+1.0, 303.3, 400.0, 239.7, 400.0, 400.0
+1.0, 133.0, 400.0, 133.0, 400.0, 400.0
+1.0, 98.4, 400.0, 117.2, 400.0, 400.0

data/SEL_dataset/test_gt/0445.txt

@@ -0,0 +1,2 @@
+1.0, 225.5, 400.0, 178.6, 400.0, 400.0
+1.0, 239.0, 400.0, 239.0, 400.0, 400.0

data/SEL_dataset/test_gt/0460.txt

@@ -0,0 +1,4 @@
+1.0, 247.6, 400.0, 285.0, 400.0, 400.0
+1.0, 205.0, 400.0, 205.0, 400.0, 400.0
+1.0, 17.3, 400.0, 303.4, 400.0, 400.0
+1.0, 159.6, 400.0, 125.7, 400.0, 400.0

data/SEL_dataset/test_gt/0461.txt

@@ -0,0 +1,3 @@
+1.0, 213.1, 400.0, 207.7, 400.0, 400.0
+1.0, 85.2, 400.0, 156.8, 400.0, 400.0
+1.0, 337.9, 400.0, 248.9, 400.0, 400.0

data/SEL_dataset/test_gt/0492.txt

@@ -0,0 +1,3 @@
+1.0, 119.6, 400.0, 85.7, 400.0, 400.0
+1.0, 214.2, 400.0, 235.3, 400.0, 400.0
+1.0, 302.5, 400.0, 390.4, 400.0, 400.0

data/SEL_dataset/test_gt/0498.txt

@@ -0,0 +1,5 @@
+1.0, 345.5, 400.0, 309.0, 400.0, 400.0
+1.0, 273.4, 400.0, 266.9, 400.0, 400.0
+1.0, 236.0, 400.0, 236.0, 400.0, 400.0
+1.0, 192.0, 400.0, 192.0, 400.0, 400.0
+1.0, 119.5, 400.0, 152.4, 400.0, 400.0

data/SEL_dataset/test_gt/0499.txt

@@ -0,0 +1,2 @@
+1.0, 267.0, 400.0, 267.0, 400.0, 400.0
+1.0, 286.0, 400.0, 290.4, 400.0, 400.0

data/SEL_dataset/test_gt/0500.txt

@@ -0,0 +1 @@
+1.0, 299.0, 400.0, 299.0, 400.0, 400.0

data/SEL_dataset/test_gt/0505.txt

@@ -0,0 +1 @@
+1.0, 215.0, 400.0, 215.0, 400.0, 400.0

data/SEL_dataset/test_gt/0511.txt

@@ -0,0 +1 @@
+1.0, 300.6, 400.0, 310.0, 400.0, 400.0

data/SEL_dataset/test_gt/0519.txt

@@ -0,0 +1,2 @@
+1.0, 254.0, 400.0, 216.7, 400.0, 400.0
+1.0, 308.9, 400.0, 276.2, 400.0, 400.0

data/SEL_dataset/test_gt/0538.txt

@@ -0,0 +1,2 @@
+1.0, 305.9, 400.0, 134.9, 400.0, 400.0
+1.0, 353.9, 400.0, 181.0, 400.0, 400.0

data/SEL_dataset/test_gt/0541.txt

@@ -0,0 +1 @@
+1.0, 276.1, 299.7, 400.0, 400.0, 400.0

data/SEL_dataset/test_gt/0548.txt

@@ -0,0 +1 @@
+1.0, 178.5, 400.0, 219.4, 400.0, 400.0

data/SEL_dataset/test_gt/0588.txt

@@ -0,0 +1 @@
+1.0, 309.3, 400.0, 289.6, 400.0, 400.0

data/SEL_dataset/test_gt/0603.txt

@@ -0,0 +1 @@
+1.0, 154.6, 400.0, 260.9, 400.0, 400.0

data/SEL_dataset/test_gt/0609.txt

@@ -0,0 +1 @@
+1.0, 319.3, 400.0, 85.3, 400.0, 400.0

data/SEL_dataset/test_gt/0625.txt

@@ -0,0 +1,2 @@
+1.0, 194.1, 400.0, 187.9, 400.0, 400.0
+1.0, 107.3, 400.0, 149.9, 400.0, 400.0

data/SEL_dataset/test_gt/0628.txt

@@ -0,0 +1 @@
+1.0, 136.3, 400.0, 95.1, 400.0, 400.0

data/SEL_dataset/test_gt/0638.txt

@@ -0,0 +1 @@
+1.0, 353.9, 400.0, 360.2, 400.0, 400.0

data/SEL_dataset/test_gt/0639.txt

@@ -0,0 +1,2 @@
+1.0, 302.9, 400.0, 308.2, 400.0, 400.0
+1.0, 333.0, 400.0, 339.3, 400.0, 400.0

data/SEL_dataset/test_gt/0640.txt

@@ -0,0 +1,2 @@
+1.0, 155.5, 400.0, 136.5, 400.0, 400.0
+1.0, 89.0, 400.0, 121.0, 400.0, 400.0

data/SEL_dataset/test_gt/0646.txt

@@ -0,0 +1,3 @@
+1.0, 215.1, 400.0, 202.7, 400.0, 400.0
+1.0, 253.0, 400.0, 365.3, 400.0, 400.0
+1.0, 111.3, 400.0, 265.7, 400.0, 400.0

data/SEL_dataset/test_gt/0647.txt

@@ -0,0 +1 @@
+1.0, 159.1, 400.0, 155.9, 400.0, 400.0

data/SEL_dataset/test_gt/0648.txt

@@ -0,0 +1 @@
+1.0, 297.7, 400.0, 277.4, 400.0, 400.0

data/SEL_dataset/test_gt/0653.txt

@@ -0,0 +1,3 @@
+1.0, 215.1, 400.0, 210.5, 400.0, 400.0
+1.0, 341.6, 400.0, 281.3, 400.0, 400.0
+1.0, 76.7, 400.0, 164.0, 400.0, 400.0

data/SEL_dataset/test_gt/0659.txt

@@ -0,0 +1 @@
+1.0, 207.0, 400.0, 207.0, 400.0, 400.0

data/SEL_dataset/test_gt/0684.txt

@@ -0,0 +1 @@
+1.0, 263.5, 400.0, 281.8, 400.0, 400.0

data/SEL_dataset/test_gt/0685.txt

@@ -0,0 +1 @@
+1.0, 291.0, 400.0, 291.0, 400.0, 400.0

data/SEL_dataset/test_gt/0691.txt

@@ -0,0 +1 @@
+1.0, 188.0, 400.0, 188.0, 400.0, 400.0

data/SEL_dataset/test_gt/0692.txt

@@ -0,0 +1 @@
+1.0, 176.6, 400.0, 199.3, 400.0, 400.0

data/SEL_dataset/test_gt/0702.txt

@@ -0,0 +1,2 @@
+1.0, 105.9, 400.0, 316.4, 400.0, 400.0
+1.0, 198.3, 400.0, 186.9, 400.0, 400.0

data/SEL_dataset/test_gt/0710.txt

@@ -0,0 +1 @@
+400.0, 389.1, 36.1, 1.0, 400.0, 400.0

data/SEL_dataset/test_gt/0730.txt

@@ -0,0 +1,2 @@
+1.0, 269.0, 400.0, 269.0, 400.0, 400.0
+1.0, 246.1, 400.0, 242.6, 400.0, 400.0

data/SEL_dataset/test_gt/0743.txt

@@ -0,0 +1,2 @@
+1.0, 340.8, 400.0, 102.6, 400.0, 400.0
+1.0, 326.3, 400.0, 239.3, 400.0, 400.0

data/SEL_dataset/test_gt/0745.txt

@@ -0,0 +1 @@
+1.0, 351.9, 400.0, 361.1, 400.0, 400.0

data/SEL_dataset/test_gt/0756.txt

@@ -0,0 +1 @@
+1.0, 359.9, 400.0, 364.1, 400.0, 400.0

data/SEL_dataset/test_gt/0764.txt

@@ -0,0 +1 @@
+1.0, 97.1, 400.0, 89.6, 400.0, 400.0

data/SEL_dataset/test_gt/0769.txt

@@ -0,0 +1,3 @@
+1.0, 279.3, 400.0, 262.4, 400.0, 400.0
+1.0, 215.0, 400.0, 215.0, 400.0, 400.0
+1.0, 85.4, 400.0, 62.9, 400.0, 400.0

data/SEL_dataset/test_gt/0774.txt

@@ -0,0 +1,2 @@
+1.0, 186.0, 400.0, 186.0, 400.0, 400.0
+1.0, 100.6, 400.0, 147.3, 400.0, 400.0

data/SEL_dataset/test_gt/0787.txt

@@ -0,0 +1,3 @@
+1.0, 227.0, 400.0, 328.7, 400.0, 400.0
+1.0, 277.5, 169.0, 1.0, 400.0, 400.0
+1.0, 102.1, 400.0, 390.0, 400.0, 400.0

data/SEL_dataset/test_gt/0788.txt

@@ -0,0 +1,2 @@
+1.0, 224.6, 400.0, 236.4, 400.0, 400.0
+1.0, 147.0, 400.0, 147.0, 400.0, 400.0

data/SEL_dataset/test_gt/0789.txt

@@ -0,0 +1,2 @@
+1.0, 297.4, 400.0, 48.9, 400.0, 400.0
+1.0, 348.4, 346.1, 400.0, 400.0, 400.0

data/SEL_dataset/test_gt/0793.txt

@@ -0,0 +1,2 @@
+1.0, 69.3, 347.6, 400.0, 400.0, 400.0
+1.0, 95.4, 400.0, 127.9, 400.0, 400.0

data/SEL_dataset/test_gt/0795.txt

@@ -0,0 +1 @@
+1.0, 268.9, 400.0, 275.3, 400.0, 400.0