IFM coding challenge
See my code in sol.py, I extract the requested features to found_tags. The confidence value is mostly useless it's just the percent of pixels in the fitted rectangle that are white
| Identified Frame | Mask |
|---|---|
 |
 |
 |
 |
Out data 1:
{
"uv": [
604.4999389648438,
677.4999389648438
],
"theta": -0.0,
"p": 623.9998779296875,
"a": 24214.990478516556,
"mu": 0.7283504825511897
}
{
"uv": [
627.87939453125,
251.41232299804688
],
"theta": -89.64412689208984,
"p": 1007.1481628417969,
"a": 60946.55979316402,
"mu": 0.8904522287095048
}
{
"uv": [
194.92147827148438,
115.00743103027344
],
"theta": -86.86975860595703,
"p": 914.7424926757812,
"a": 51827.32436198741,
"mu": 0.8353122707556244
}
Out data 2:
{
"uv": [
859.8844604492188,
826.300537109375
],
"theta": -1.7899105548858643,
"p": 426.72918701171875,
"a": 11317.92199785792,
"mu": 0.9462867832122382
}
{
"uv": [
408.99993896484375,
739.4999389648438
],
"theta": -0.0,
"p": 557.9999084472656,
"a": 19117.99389648484,
"mu": 0.9592533661898452
}
{
"uv": [
906.5,
445.5
],
"theta": -0.0,
"p": 543.9999084472656,
"a": 18206.994033813942,
"mu": 0.9967591593606022
}
{
"uv": [
41.5,
391.5
],
"theta": -0.0,
"p": 440.0,
"a": 11371.0,
"mu": 0.9813560812593439
}
{
"uv": [
1331.0,
315.0
],
"theta": -0.0,
"p": 596.0,
"a": 21840.0,
"mu": 0.9322802197802198
}