Pre-requisites
It is necessary to en existing media
Use the media_id to send the GET request to the endpoint /predict/v2/media/media_id
Examples of successful responses
Images
For an existing image media, the results should look like this:
{
"media_id": "12ee330a-9142-4640-838c-6d6fc1e65ddd",
"name": "sample_ad",
"status": "done",
"type": "image",
"aoi_status": "unspecified",
"results": {
"heat": "https://files.neuronsinc.com/media_nvai_heat.jpeg",
"clarity_heat": "https://files.neuronsinc.com/media_ve_clarity_heat.jpeg",
"engagement_heat": "https://files.neuronsinc.com/media_ve_excitedness_heat.jpeg",
"fog": "https://files.neuronsinc.com/media_nvai_fog.jpeg",
"formatted": "https://files.neuronsinc.com/media_nvai_formatted.jpeg",
"first_two_heat": "https://files.neuronsinc.com/media_nvai_first2_heat.jpeg",
"first_two_fog": "https://files.neuronsinc.com/media_nvai_first2_fog.jpeg",
"last_two_heat": "https://files.neuronsinc.com/media_nvai_last2_heat.jpeg",
"last_two_fog": "https://files.neuronsinc.com/media_nvai_last2_fog.jpeg",
"engagement_new_heat": "https://files.neuronsinc.com/media_nvai_engaging_heat.jpeg",
"memory_heat": "https://files.neuronsinc.com/media_nvai_memory_heat.jpeg",
"buy_heat": "https://files.neuronsinc.com/media_nvai_buy_heat.jpeg",
"trustworthy_heat": "https://files.neuronsinc.com/media_nvai_trustworthy_heat.jpeg",
"annoying_heat": "https://files.neuronsinc.com/media_nvai_annoying_heat.jpeg"
},
"metrics": {
"cognitive_demand": 50.904459874777665,
"focus": 83.45828370672884,
"focus_first_two": 80.62319981104667,
"focus_last_two": 45.30343438115154,
"clarity": 63.41,
"engagement": 67.51,
"engagement_new": 58.3116969754619,
"memory": 56.13035924973027,
"buy": 73.59597183042958,
"trustworthy": 46.642828372217,
"annoying": 22.09484428167343
},
"created_at": "2023-03-01T09:54:49Z",
"updated_at": "2023-03-01T09:54:49Z"
}
NOTE: The assets will only be available under the URL's for a maximum period of 5 minutes, so make sure to download them before they expire.
Image prediction results
Below there's a brief explanation of each field in the image prediction results:
- media_id: The unique identifier, which can be used to query the media.
- name: The name of the image file.
- status: The prediction status. If status is done then results should be visible in the response.
- type: The type of media (image or video).
The heat maps are stored as image assets (with an expiry, as explained above), and links can be find under results:
- heat: This is the attention heat map.
- clarity_heat: This is the heat map for clarity.
- engagement_heat: This is the engagement heat map.
- fog: This is the attention fog map.
- formatted: This is the original image (no heat map).
- first_two_heat: This is heat map shows what draws attention in the first two seconds.
- first_two_fog: This is fog map shows what draws attention in the first two seconds.
- last_two_heat: This is heat map shows what draws attention in the final two seconds.
- last_two_fog: This is fog map shows what draws attention in the final two seconds.
- engagement_new_heat: The heat map for the brand new engagement metric.
- memory_heat: The heat map for the brand new memory metric.
- buy_heat: The heat map for the brand new intent to buy metric.
- trustworthy_heat: The heat map for the brand new trustworthiness metric.
- annoying_heat: The heat map for the brand new annoyance metric.
Metrics (find out what each metric means in this article):
- cognitive_demand: The value for cognitive demand (in percentage).
- focus: The value for total attention (in percentage).
- focus_first_two: The value for attention during the first two seconds (in percentage).
- focus_last_two: The value for attention during the final two seconds (in percentage).
- clarity: The value for clarity (in percentage).
- engagement: The value for engagement (in percentage).
- engagement_new: The value for the brand new engagement metric (in percentage).
- memory: The value for the brand new memory metric (in percentage).
- buy: The value for the brand new intent to buy metric (in percentage).
- trustworthy: The value for the brand new trustworthiness metric (in percentage).
- annoying: The value for the brand new annoyance metric (in percentage).
Videos
For an existing video media, the results should look like this:
{
"media_id": "0d1aeafa-d360-47c2-afaf-c6149281c202",
"name": "video_media",
"status": "done",
"type": "video",
"aoi_status": "done",
"results": {
"heat": "https://files.neuronsinc.com/media_nvai_heat.mp4",
"fog": "https://files.neuronsinc.com/media_nvai_fog.mp4",
"formatted": "https://files.neuronsinc.com/media_nvai_formatted.mp4",
"metrics": "https://files.neuronsinc.com/media_nvai_metrics.json",
"frt_frame_metrics": "https://files.neuronsinc.com/media_frt_metrics.json",
"brand_detection": "https://files.neuronsinc.com/nv-test/media_brand_detection.json?",
"brand_detection_metrics": "https://files.neuronsinc.com/media_brand_detection_metrics.json?",
"aoi_scores": "https://files.neuronsinc.com/media_aoi_metrics.csv"
},
"metrics": {
"cognitive_demand": 91.73756099979508,
"focus": 39.50367547395666,
"focus_first_two": 68.92043509687068,
"focus_last_two": 12.776237332510023,
"engagement_new": 48.56829226514142,
"engagement_new_start": 60.587450885026215,
"engagement_new_end": 48.31150801571462,
"memory": 56.917629571266,
"memory_start": 51.441789723350155,
"memory_end": 52.40626494502513,
"buy": 63.22354314204671,
"buy_start": 52.2297100223787,
"buy_end": 36.479611806214365,
"trustworthy": 77.95613007510966,
"trustworthy_start": 63.80634240907102,
"trustworthy_end": 52.42010053947325,
"annoying": 26.19246212057231,
"annoying_start": 28.379557578412584,
"annoying_end": 28.379557578412584
},
"created_at": "2024-01-19T09:05:44Z",
"updated_at": "2024-01-19T09:36:49Z"
}
Where the brand detection metrics can be found in the JSON indicated in the brand_detection_metrics field, the AOI detection results can be found in the JSON file indicated in the aoi_scores field.
Video prediction results
Below there's a brief explanation of each field in the video prediction results:
- media_id: The unique identifier, which can be used to query the media.
- name: The name of the video file.
- status: The prediction status. If status is done then results should be visible in the response.
- type: The type of media (image or video).
- aoi_status: The status of the AOI prediction for the media.
The heat maps are stored as video assets (with an expiry, as explained above), and links can be find under results:
- heat: This is video with its attention heat map.
- fog: This is the video with its attention fog map.
- formatted: This is the original video (no heat map).
Additionally, the results field contains some JSON assets with additional detection metrics:
- metrics: This JSON contains the frame-by-frame prediction metrics.
- frt_frame_metrics: This JSON contains the frame-by-frame FRT metrics.
- brand_detection: This JSON contains the frame-by-frame information on detected areas-of-interest (AOIs) that resulted from the brand-detection.
- brand_detection_metrics: This JSON file contains the aggregated result for each of the detected AOIs prediction metrics.
- aoi_scores: This file contains contains the brand detection metrics in a CSV format.
Metrics (find out what each metric means in this article):
- cognitive_demand: The value for cognitive demand (in percentage).
- focus: The value for attention (in percentage).
- focus_first_two: The value for attention during the first two seconds (in percentage).
- focus_last_two: The value for attention during the final two seconds (in percentage).
- engagement_new: The value for the brand new engagement metric (in percentage).
- engagement_new_start: The value for the brand new engagement metric during the first two seconds (in percentage).
- engagement_new_end: The value for the brand new engagement metric during the last two seconds (in percentage).
- memory: The value for the brand new memory metric (in percentage).
- memory_start: The value for the brand new memory metric during the first two seconds (in percentage).
- memory_end: The value for the brand new memory metric during the last two seconds (in percentage).
- buy: The value for the brand new intent to buy metric (in percentage).
- buy_start: The value for the brand new intent to buy metric during the first two seconds (in percentage).
- buy_end: The value for the brand new intent to buy metric during the last two seconds (in percentage).
- trustworthy: The value for the brand new trustworthiness metric (in percentage).
- trustworthy_start: The value for the brand new trustworthiness metric during the first two seconds (in percentage).
- trustworthy_end: The value for the brand new trustworthiness metric during the last two seconds (in percentage).
- annoying: The value for the brand new annoyance metric (in percentage).
- annoying_start: The value for the brand new annoyance metric during the first two seconds (in percentage).
- annoying_end: The value for the brand new annoyance metric during the last two seconds (in percentage).