Results

Here's an example of the result of an analysis performed by Foyer Insight:

{
"classifier": [
{
"confidence": 0.9996496438980103,
"name": "Kitchen",
"rank": 1
},
{
"confidence": 0.9999802112579346,
"name": "Indoor",
"rank": 2
}
],
"detections": [
{
"area": 55518,
"attributes": {
"is-stainless": false
},
"bbox": [
3,
156,
564,
356
],
"class": 164,
"classId": 164,
"score": 0.8126423358917236
},
{
"area": 66851,
"attributes": {
"is-stainless": false
},
"bbox": [
0,
3,
540,
271
],
"class": 86,
"classId": 86,
"score": 0.7137435674667358
}
...
]
}

Let's break down the sections we're seeing.

Classifier

Each classifier has three fields: confidence, name, and rank.

  • Confidence is how certain Foyer Insight is of the classification. This is a number between 0 and 1, which alternatively can be thought of as a percentage. In this case, Insight is almost certain we've provided an image of an indoor kitchen.

  • Name is the full name of the image's classification.

  • Rank is a number greater than zero. The classification with rank 1 is always the one with the highest confidence.

Detections

Each detection has six fields: area, attributes, bbox, class, classId, and score. A detection is an individual object or group of related objects that Insight found in the image.

  • Area is the total number of pixels a detection takes up in the entire image. For example, given the above result, 55518 pixels in the image are detected as being part of a table.

  • Attributes are extra details about the detection. In our result, no objects were detected as being made of stainless steel.

  • BBox is the bounding box of the detection. It is made up of 4 numbers, which are really 2 (x, y) coordinate pairs representing the upper left and bottom right corners respectively. For example, the values [3, 156, 564, 356] represent a point at (3, 156) and (564, 356).

  • Class is the full name of the detection's class.

  • ClassID is the integer ID that corresponds to a particular class name.

  • Score is how certain Insight is of the class for a detection. As with a classifier's score, it is a number between 0 and 1. In our example, Insight is about 81% certain that its detection of a table is correct.