Frequently Asked Questions

If you have any questions about our company, products or our policies you do not need to wait for an email response or for a scheduled call with one of our representatives. All you need to do is to head below and browse the frequently asked questions that we have made in order to answer your most commonly asked questions about us and our products.

1. General Questions about Sightcorp Privacy Policy

When using the website, various cookies are saved on your devices, which can then be accessed. Cookies are small pieces of information (in the form of text) that a server sends to the browser (such as Internet Explorer or Firefox) with the intention that the browser sends this information back to the server the next time a user makes use of the website. Cookies cannot damage your Devices or the files saved on it.

For more general information on cookies see the Wikipedia article on HTTP Cookies.

This paragraph applies to those cookies that are stored and read by Sightcorp itself. No third parties are involved in storing and reading these so-called first party cookies. Sightcorp only uses first party technical and functional cookies. Such cookies are used by Sightcorp to show you our website, to make it function correctly, to remember your preferences and settings (as username, password, language etc.) and to help you to use the website efficiently and effectively. Technical cookies are necessary for our website to function properly.

Functional cookies are not strictly necessary for the functioning of our website, but they add functionality for you and enhance your experience.

By using the website you can get access to social media websites. Using these so-called social plug-ins you can log through your social media profile and subsequently share information from the website with others. With these social plug-ins third party cookies are saved on your Devices. These cookies serve the purpose of optimizing your user experience.

Sightcorp uses analytic cookies to collect information about and analyze the use of the website, in order to make improvements.

Sightcorp uses Google Analytics in order to collect statistics about the use and visits of the website. With these analytical cookies a permanent cookie is saved on your Device to register your use of the website. Google Analytics analyzes this data about the website and provides Sightcorp with the results. This way, Sightcorp obtains insight into the traffic of the website and the way in which the website is used. Based on this information Sightcorp is able to make adjustments to the website. The information that Sightcorp obtains is transferred to Google and stored by Google on servers in the United States. Sightcorp has taken the following measures to protect  your privacy:

  • Sightcorp has entered into a data processing agreement with Google in which agreements have been made regarding the processing of personal data.
  • Sightcorp does not allow Google to use the obtained data for other Google services.
  • Sightcorp has removed the last octet of your IP address. An IP address consists of 4 so-called octets of 3 digits each. Google offers the opportunity to remove the last octet of the IP address. This happens before your IP address is retained by Google.
  • Sightcorp does not make use of other Google services in combination with Google Analytics.

2. General Questions About Sightcorp Products

Of course you can. You can get a free trial for 14 days of any of our products. All you need to do is to contact us and have a call with one of our colleagues. You will have access to the full version of our software, where the only limitation is time - 14 days.

Facial landmarks are used by our software in order to localize and represent salient regions of the face. The example of such Mask Points are: Eyes, Eyebrows, Nose and Mouth.

Our DeepSight software licensing is different for the plug-and-play Toolkit application and the SDK.

The DeepSight Toolkit is licensed under a monthly subscription model* with a minimum period of 1 year and a minimum order of 3 licenses. Toolkit license prices are discounted based on volume and contract duration (2-5 years). New customers are required to purchase a one-time setup fee which includes 16 hours of support from our team to help you get familiar with the software. The Toolkit can be tried with a free trial license valid for 14 days.

* A life-time license fee model is available on request, contact our sales team (link)

For DeepSight SDK the licensing is adapted to the different stages of your software solution development cycle. The first step is to start with a developer license to program your application or to integrate the DeepSight SDK into existing platforms. The developer license is a time-based (e.g. 3, 6, or 12 months) and device-based license. When you're ready to launch your product application you will need a production release license, for which there are two license models available:

  1. A lifetime release license with volume discounts and a yearly maintenance and support fee
  2. A time-based release license with a minimum order and license period

Both the development and production release licenses offer e-mail & phone support, technical documentation, software code samples and access to new releases of the DeepSight SDK. The SDK can be tested with a free trial license valid for 30 days.

As we work with tiered pricing, we suggest you contact our sales department for a quotation. If you still have doubts regarding licensing, do not hesitate to contact our sales team.

Yes. You can contact us by clicking here.

Our F.A.C.E. web API provides face analysis as a service including the analysis of emotions. Our CrowdStats platform provides demographic reporting, including 'mood' time-series reports.

Sightcorp as a company does not provide video analysis as a service however, our CrowdSight Toolkit can be used to analyze faces in a video. Normally, the application carries out a real-time analysis via camera recording however users can also upload a pre-existing video for the same type of analysis. The Toolkit is most effectively used in combination with the CrowdStats reporting dashboard which displays all analyzed data. 

SDK stands for Software Development Kit. This means that it is a technology toolbox (library) which can be used by programmers to integrate into new or existing applications.

The SDK is designed for Human-Computer Interaction and Human Behavior Analysis. We consider the SDK to be an enabling toolbox for many applications; it is up to you to find great ideas on how you can use these technologies.

If you couldn’t find the answer to your question, click here to submit your support request form.

3. Trial Questions

Product trials can be requested on the product pages themselves by selecting the 'Request Trial' or 'Contact Us' buttons.  Additionally, you can request a trial by filling out a trial request form in the Contact section of the website. Or by contacting the sales department at sales@sightcorp.com.

Trials are provided for a period of 14 days. If you require further testing after this duration one of our friendly business development team will assist you and discuss your individual requirements. 

All product trials that we provide are free of charge. We want to give you the opportunity to not only experience the software but to also see its value for yourself.  

Our dedicated team of business development specialists are on hand to discuss your individual use cases for the face analysis products. Arranging an introductory call is the best place to start for guidance on which product is right for you. Contact them via email, sales@sightcorp.com or by filling out a contact request form.  

Yes, it is possible to trial multiple products at once.

4. DeepSight General Questions

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).

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. 

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.

During the development phase, the DeepSight products need internet connection every time you start up the program. When you enter the redistribution phase with a release license, the SDKs need one-time internet connection for license activation.

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.

The F.A.C.E. API does require an internet connection for every execution since it is a Cloud based application.

At Sightcorp, we aim to make our computer vision products affordable and easy to use for all our customers. However, we do not publish the product pricing due to variation in price based on volume, duration and license type per product.

Please fill out a request form for your desired product, and we will get back to you shortly with a quote.

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).

Deep learning is a subset of machine learning and a specific approach used for the building and training of neural networks. A deep learning model is designed to continually analyze data with a logic structure similar to how a human would draw conclusions. To achieve this, deep learning uses a layered structure of algorithms called an artificial neural network (ANN).

Deep learning differs from traditional machine learning techniques in that the input data is passed through a series of nonlinearities or nonlinear transformations before it becomes output. To further explain, a machine learning algorithm often needs an engineer to step in and make adjustments if it returns with an inaccurate prediction. With a deep learning model, however, the algorithms can determine on their own if a prediction is accurate or not.

DeepSight technology is a machine learning-based best-in-class and highly accurate Face Analysis software able to detect faces from any angle and distance, and in real time. Face Detection, Age and Gender Estimation are the core features of our proprietary deep learning algorithms. Additionally, it includes Mood and Happiness detection capabilities for Retail and Digital Signage applications.

Sightcorp has a standing commitment to protect the customer's and audience’s privacy. All of our products are fully respectful and complaint with privacy regulations: data is collected anonymously, no images are recorded or stored therefore no uniquely identifiable data can be extracted. Images from the camera sensors undergo real-time conversion into data without any storage. For this reason, you can still perform all face analysis tasks with the additional capability to prevent others from being able to view video streams or see people’s faces thanks to our face blur solutions. 

DeepSight (Toolkit) analyzes emotions based on smile. The smile is represented in a value between 0-100 and this metric allows you to define a person's mood as neutralhappy or very happyIt can be used as a customer satisfaction index, for example in retail environments around the (PoS) payment area. The smile detector is a very robust data metric that can be accurately measured in any real-life environment and under changing (lighting) conditions. 

Using DeepSight, 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.

DeepSight features are:

  • People Counting
  • Face Detection
  • Face Tracking and Counting
  • Multi-People Analysis
  • Age Estimation
  • Gender Estimation
  • Mood Recognition
  • Face Blurring
  • Attention Time
  • Face Mask Detection

Yes, it can. People counting is a technology used to measure the number of people visiting a location and uses face detection technology as the counting method. This method is completely privacy compliant due to our automatic face blur feature incorporated in DeepSight.

People counting software quantifies activities in a physical space. It determines where visitors are, the time they spend at a particular location, and the time they arrive.  

5. DeepSight COVID-19 Questions

No. The occupancy monitoring system has been designed only to capture footfall inside the building.

Occupancy counting has long-term benefits that will extend past its use as a safety tool during COVID-19. Footfall monitoring provides useful insights into how spaces are being used and help to potentially reduce unnecessary rental costs. Additionally, crowd analysis software can be used to identify peak hours for retail businesses, helping them to create better staffing plans in accordance with peak hours and shop demands.

The design of your building comes into play when it comes to selecting where the best place is to position your digital screens. In most cases, building entrances seem to be the prime location as they can accurately monitor traffic coming into the building and ensures that people can clearly see occupancy updates as they approach and don’t know if it is ok to enter.

Our occupancy monitoring system can be implemented and installed in any type of building to keep track of footfall and capacity limits.The product is easily integrated and discrete, which therefore makes it practical for any setting. Due to the Covid-19 crisis, our team has designed our crowd analysis system around buildings reopening safely and responsibly with privacy at the forefront of our product design.

No. Our product has been designed with ease of integration in mind, meaning that it is simple to install with no in-depth technical knowledge required to operate it. The product also comes with easy to follow instructions which acts as a useful reference guide.

No. Information pertaining to individual identity is never gathered or stored. Instead, data is collected anonymously and in an encrypted format.

6. DeepSight Toolkit Technical Questions

With the DeepSight toolkit you have the option to store data locally on disk in CSV format. You can choose to write raw data and/or aggregated data.

As the user of the Toolkit, you are in full control of the data. The CSV files are stored in a user-defined location.

The Toolkit allows easy integration into any platform using its web interface to communicate with GET/POST requests in JSON format.

Using the web interface, you can make GET requests to the Toolkit to retrieve the data or set an URL in the Toolkit to send POST requests. The Toolkit is agnostic to the programming language, you can use your favorite language to process the requests.

The DeepSight Toolkit has been designed to be user friendly, therefore it can be easily operated with no existing programming knowledge. 

The DeepSight Toolkit software can run on Windows or Linux, depending on your preferences. 

6. DeepSight SDK Technical Questions

  • The SDK does not store any personal or privacy sensitive information. Image data is being discarded once the faces have been analyzed, and the SDK returns a structure that contains anonymous demographic data. 

Since the SDK doesn't store any data, the flow of data outside of the SDK is up to the user.

You can use the SDK to do a direct integration with C++ or wrap it in another language using an interfacing language, for example Java/JNI or C#/CLI. We do not provide wrappers to these languages ourselves.

DeepSight SDK is written in C++. You can use the SDK on Windows using Visual Studio, or on Ubuntu using GCC. The SDK also comes with a standalone service, which allows the user to use POST requests using their favorite programming language.

Our SDK works on x86, x86-64 and ARM based hardware as long as the OS is supported : Windows ( x86, x86-64 ), OSX ( x86-64 ), Linux ( x86, x86-64, ARM ), Android ( ARM ), iOS ( ARM ).

Officially supported : Windows ( x86 ), OSX ( x86-64 ), Ubuntu ( x86-64, ARM ), Android ( ARM ), iOS ( ARM ).

A full HD web camera (eg: Logitech HD pro series) is required for enhanced tracking accuracy. 

Moreover, depending on your camera manufacturer, it should be possible to configure camera parameters for best results. The most important this is to enable the anti-flicker for your region (50 or 60 Hrz) and to disable the automatic gain, white balance and focus control. This is a fundamental step, as the software will recognize any changes in the images as a change in the face, significantly affecting the classification and tracking results.

The faces must be sufficiently and frontally lit. In case of nonlinear light patterns on the face or strong sideways illumination, DeepSight SDK will have more difficulty to correctly process the features of a person. Back illumination, such as a window behind a person, darkens the face feature and is thus less desirable. This can be counteracted to some extend by disabling automatic camera adjustment such as white-balance, focus and exposure controls, however this requires that the camera position or illumination does not change significantly. 

The further the person is from the camera, the smaller the resolution of the face in the image will be. DeepSight SDK performs optimally with faces of 150x150 pixels. The minimum face size for DeepSight SDK to perform in a meaningful way is 50x50 pixels (but the accuracy is lower).
 
In order to detect people from a distance, two things are required. Firstly, the capture resolution of the camera must be sufficiently large, 640x480 minimum is advised and 1080p is recommended. Secondly, optical zoom can be used in order to detect faces with sufficient resolution. As a general rule of thumb, when trying to record people from a distance larger then 5m you would require a camera with a zoom lens.

The ideal positioning of the camera would be aimed at a narrow space where people pass one by one, such as a doorway. The reason for this is that you can zoom in on say upper part of the door and capture each person's face one by one while they are entering or exiting. This allows for a higher face image resolution as opposed to the scenario where people can pass by alongside each other and thus a lesser level of zoom must be applied to process multiple people at once.
 
The best results are obtained when people are captured frontally. We advise to keep the vertical angle between the face and the camera no more than 15 degrees. If you want to detect people at extreme angles, such as from cameras that are hanging on the ceiling you could attempt to attract their attention,for example by installing the camera next to an advertisement screen.

7. Face Mask Detection General Questions

Our mask detection model performs with a 98% accuracy rate (tested on a reduced MAFA dataset).

Our face mask detection technology uses a combination of the latest face detection and face analysis computer vision algorithms to detect and analyze faces with and without masks.

Yes, our face mask detection can work with existing USB, IP and CCTV cameras in real-time.

Yes, face mask detection offers anonymized analysis which is completely privacy-proof: no user personal data is retrieved or stored as the system anonymously checks facemasks only.

Yes, the software outputs a confidence value for each detection in the raw data output. Users are able to set their own thresholds according to their specific scenario. The software also outputs a yes/no value for each ID in the aggregated data output which uses a predefined threshold.

Yes, we offer 2-week free trials for all of our products. Contact us for more details.

DeepSight with face mask detection is running locally on the edge device and all data is processed and stored on the same device by default. The software does not store any personal data, only information about face masks.

Face masks detection is a part of our existing DeepSight products which are ready-to-use solutions.

Face mask detection has the same minimum requirements as our DeepSight products. Find more information about preferred devices here.

Did You Find An Answer To Your Question?

If you still have questions please do not hesitate and contact us, we will be glad to answer any questions that you might have about us or our products!