Create a JSON file with the following params:
{
"fps": float, // Greater than 0.
"width": integer, // # columns, max. 30.
"height": integer, // # rows, max. 30.
"low": float, // Min. value range.
"high": float, // Max. value range, low < high.
"nFrames": integer, // Number of frames.
"frames": [
[f1_x1_y1, f1_x2_y1, ..., f1_x1_y2, f1_x2_y2, ... f1_xw_yh],
[f2_x1_y1, f2_x2_y1, ..., f2_x1_y2, f2_x2_y2, ... f2_xw_yh],
...,
[fn_x1_y1, fn_x2_y1, ..., fn_x1_y2, fn_x2_y2, ... fn_xw_yh]
]
}
Here you have a brief explanation about these params:
FPS: Frame rate of the video or data where you extracted information to create this surface. This value must be greater than 0.
Width: Number of columns of the surface. Columns are related to the length of the field, i.e. around 105 metres in football. This value must be greater than 0 and less or equal than 30.
Height: Number of rows of the surface. Rows are related to the width of the field, i.e. around 68 metres in football. This value must be greater than 0 and less or equal than 30.
Low: Lowest value the surface could have in a cell.
nFrames: Number of frames that contains the array "frames".
High: Highest value the surface could reach in a cell. This value must be greater than the low one.
Frames: Array where each frame is another array in row-major order that describes the surface values. First element of each frame array (first row and column) points to the top-left corner of the field. Length of each frame array must be equal to the multiplication of values width and height.
Example:
{
"fps": 25.0,
"width": 2,
"height": 2,
"low": 0.0,
"high": 1.0,
"nFrames": 1,
"frames": [
[0.1, 0.1, 0.9, 0.9]
]
}
Once you've created the .json file, go to cloud.metrica-sports.com
You can then click on the 'more options' icon from a Video Project.
Then click on the Import icon next to Surfaces row to upload the json file: