What is DeepSight Toolkit?
Using DeepSight Toolkit, you gain accurate insights into your audiences in real-world environments. Anonymously detect & analyze faces, discovering insightful data on demographics, behavior patterns, interests, satisfaction levels and much more.
Empower your business or brand with DeepSight Toolkit now and make informed data-driven decisions to maximize your ROI.
- People Counting
- Age, gender, and head pose estimation
- Smile detection
- Viewers vs. Impressions
- Attention vs Dwell Time
- Privacy by default: face blurring
- Data processed offline and locally
- Raw & aggregated JSON and CSV data output
- Data shareable with multiple endpoints
- Ready CMS integrations
- Windows or Linux
- Minimum Requirements:
- CPU Intel i3 /i5/ i7 at least 6th generation
- Video input resolution of 640×480 (min)
- RAM: 4GB minimum
- Storage: at least 2Gb free space
- Camera: 1280 x 720 (recommended)
Audience Analytics for Real-World Environments
Detect and anonymously analyze individual faces in images or video. Understand human behaviors and gain valuable insights such as age, gender, mood, attention and other interesting metrics.
Carry out crowd analysis using our people/body counting technology. In challenging scenarios, even in areas with large, unpredictable crowds or places where occlusions are a possibility, our technology remains effective and highly accurate.
Attention tracking. Accurately measure your audience reach on your DOOH network. Count the number of impressions and true viewers and measure their attention and dwell time. Use this data to sell ad space at a premium/higher CPM.
Audience Data Reporting for DOOH
Collect historical data: Understand the number of people passing a location and their demographic breakdowns to create a qualified overview that measures audience reach for advertisers.
Audience data for each campaign: Data is collected in real-time as content is being displayed, combine this data with proof of play from the CMS to understand ad performance metrics (OTS, Impressions, views, demographics) and trends, perform extensive A/B testing exercises to maximize viewability and capture rates in target segmented audiences.
Dynamic content Triggering: Additional to reporting, you can also use the real-time audience data for dynamic triggering.
Data is delivered to the analytics visualizer in real-time and is easily accessible for around the clock insights. Viewer metrics include:
Performance Metrics for In-store Analytics
Improve your in-store marketing strategies, customer satisfaction levels and boost conversion rates using our anonymous face analysis solution.
Footfall Counting: Track and count people entering your store for accurate data about your daily visitors. Manage occupancy levels, optimize staff allocation and benchmark your stores using the data.
Demographic Breakdowns: Using demographic analysis, you can gain insightful data into the main demographic groups visiting your store and at which times they visit. Analyze metrics such as age and gender and market your products accordingly.
Sentiment Analysis: Discover how your customers are responding to your store and staff. Using mood detection, you can accurately measure the satisfaction of your customers using their genuine responses.
Dynamic Content Triggering
With the integration of the Toolkit with a CMS you can change content dynamically. Anonymously analyze faces using deep learning technology to detect individuals in and around digital screens.
Dependent on who is in front of the screen, tailored content will play suited to the demographic of the individual present. If there is a group detected around the screen, the rules defined in the CMS will decide which content to play according to the dominant demographic present at the time.
All In One Industry Solution
Using DeepSight Toolkit, you can gain accurate insights into your audiences in real-world environments. The extensive amount of analysis features available such as people count, age estimation, gender analysis and its ease of integration makes DeepSight Toolkit the perfect AI software to implement across any industry or application field.
Privacy By Design
The DeepSight Toolkit has been carefully designed with privacy at its core. Our face analysis product is completely GDPR compliant, which means that no personally identifiable information is ever collected or stored, and that it also features an extra level of protection with our automatic face blurring application. Face blur ensures individual anonymity by blurring the faces of anyone featured in live video streams in real-time. Therefore, with DeepSight you can gain accurate insights with ultimate peace of mind.
“By combining Intuiface, best-of-breed for dynamic content delivery and analytics, with Sightcorp, best-of-breed for AI-powered real-time face analysis, we have cracked the code and created something that can be adopted by any team interested in audience-aware, data-rich DOOH deployments.”
– Vincent Encontre, COO, Intuiface
What is DeepSight Toolkit?
DeepSight Toolkit is an easy-to-use cross-platform solution that enables you to detect and analyze multiple faces using just one application. The software is the deep learning-based successor to CrowdSight Toolkit, which gives you accurate insights into the behavior of your audience in real-world environments. No hard coding is required, which means that anyone can install and use the software within minutes.
How can you visualize your data?
The Toolkit outputs raw and aggregated data in JSON or CSV formats. The raw data can be used for custom data analysis, while the aggregated output is ready to be used by any user. The Toolkit can also be used to push the data directly into any BI platform so you have all your data in one place! Transform your analyzed data about viewer engagement, sentiment, attention, and audience demographics into ready-to-use graphs and charts.
Get an easy overview of the aggregated data with our complimentary Toolkit Visualizer. You will receive the Visualizer when applying for the Toolkit trial. The Visualizer shows in an easy to interpret interface of the data that was analyzed over a 24-hour period. The Toolkit also comes with a ready integration with Cloud based dashboards. Contact us for more information.
What about privacy?
Sightcorp has a standing commitment to protect the privacy of customers and audiences. Since no images are recorded or stored with our Face Analysis solutions including the Toolkit, no uniquely identifiable data (personal data) can be extracted. Images from the camera sensors undergo real-time conversion into (anonymous) statistical data within a few tenths of a second. By default, the images are then discarded immediately without any storage of facial images.
Sightcorp Face Analysis software does not store any images. The analysis process works as follows:
1. Images (frames) come in via the camera
2. Images do not get stored, but are immediately blown up into “pixels”
3. The images are thrown out/discarded
4. The only thing that is stored locally on the machine is the raw data
5. Sightcorp can’t access your data on the device or the camera
Sightcorp technology is built with privacy by design, and we have also launched a privacy by default feature. This feature activates automatic face blurring during real-time video analysis.
More features will be added soon. This will come at no extra cost, so sign up today to start getting even better results from your real-world customer analysis!
Yes. DeepSight Toolkit can analyze multiple faces at once. There is no maximum limit of faces detected in one frame but the performance of multi-face analysis depends on the power of the processor (e.g. the more people to analyze the more powerful processor is needed).
Certain degree of occlusion and tilt is acceptable, however faces in extreme angles (profile) can be difficult to analyze. The main features of the faces such as eyes, nose, mouth need to be visible for our software to be able to detect it. We can however still detect faces wearing masks.
No. There is no limit for faces however the performance of multi-face analysis depends on the power of the processor (e.g. the more people to analyze the more powerful processor is needed).
In the context of digital signage and programmatic advertising, attention time refers to a period of time captured while a person is looking at a digital billboard or sign. This can also be transferred into the context of a store, where retailers can assess how long individuals spend looking at various products. The feature is adaptable and flexible, making it a convenient product for any use case.
DeepSight runs locally on the edge and all data is processed and stored on the same device by default. The software does not store any personal data, only raw data.
The DeepSight Toolkit requires an internet connection at start-up for validating the license but it does not require internet connection for the real-time analysis.
Ready To Get Started With DeepSight Toolkit?
Test it free for 2 weeks on your favorite desktop platform