An AOI (Area of Interest) refers to elements in the asset (image or video) with a specific interest to the Neurons AI score. In order to produce scores specifically for f.x. Brand (Logo) we require a bounding box or polygon box around the target element.
AOI attention metrics
To aid the optimisation of creatives for the most impact on the market, Neurons AI offers attention metrics on AOI level. This section elucidates the technical underpinnings and visualization strategies for three pivotal metrics: Total Attention, Time Spent, and Percentage Seen. Together, they provide a comprehensive, granular, and actionable view into how consumers interact with visual content, setting the foundation for data-driven decision-making in marketing and design.
Total Attention
Total Attention on AOI is the percentage of visual attention that falls within the area for the entire exposure, which allows you to determine what draws significant attention, and what falls through the cracks. The Total Attention metric is derived from our proprietary Total Attention Heatmap algorithm. This algorithm processes raw attention data, normalizes it across different viewer demographics, and subsequently maps it onto the tested asset's AOIs. The final score represents an aggregate measure of attention, indicating the intensity and frequency with which consumers focus on each AOI during the entire exposure duration.
Total attention on AOI is available for both images and video.
Image AOI metrics
Time Spent
Time spent on AOI is the predicted amount of time customers will spend viewing the area during a 5-second exposure. The Time Spent metric utilizes our refined attention prediction algorithm, which models the temporal aspect of attention. It factors in the duration of focus, capturing moment-to-moment fluctuations in consumer gaze across the AOIs during a 5-second exposure window. The cumulative time spent on each AOI is then translated into a score, enabling you to gauge the staying power of individual elements.
Percentage Seen
Percentage Seen is the percentage of customers who will look at the area at least once during their exposure. Derived from the Total Attention Heatmap, this metric estimates the percentage of viewers likely to notice a specific AOI (Area of Interest). It uses a polynomial regression to link attention scores to the percentage seen. While CNNs predict attention maps, time spent is assessed with a linear regression. The analysis incorporates raw image data, predefined AOIs, and historical attention data to yield an output between 0-100%, indicating the expected viewer engagement with the AOI.
Beyond attention-derived metrics, we provide specific metrics derived from the score heatmaps.
Cognitive Demand on AOI: measures the cognitive demand within the area, which enables you to identify what elements are the most complex and demanding for customers to process.
Memory on AOI: measures the memorability of the specific area, which shows you which elements customers will experience as the most memorable.
Engagement on AOI: measures engagement within the area, so you can see which elements will make customers feel the most excited and immersed.
Video AOI metrics
![]()
Availability
The Availability metric utilizes our refined attention-tracking algorithm, which models the temporal aspect of attention. It factors in the duration of focus, capturing moment-to-moment fluctuations in consumer gaze across the AOIs during the exposure window. The cumulative AOI availability is then translated into a score, enabling you to gauge the staying power of individual elements in the video.
![]()
Relative Attention
Derived from the Total Attention Heatmap, this metric estimates the concentration of attention in percentage that an AOI would have when it was visible in a video.
Relative attention is calculated by dividing the sum of the attention the AOI receives, by the number of frames where the AOI was detected. This metric provides another perspective to AOI-level attention enabling users to ensure that when the AOI is visible, it receives enough attention to be impactful.
Start Attention on AOI: The percentage of visual attention that falls within the AOI for the first 2 seconds of an exposure, allowing you to identify whether the element draws immediate customer attention.
End Attention on AOI: The percentage of visual attention that falls within the AOI for the last 2 seconds of an exposure, letting you identify whether the element is in final focus.