Services

×

Leverage our proactive R&D to unlock new efficiencies

Our research and development department unites over 20 result-driven experts with hands-on cross-domain experience. We’re constantly working on internal projects to try out innovative concepts and approaches able to cut costs and minimize time to market for your product.

R&D Projects Productization

Our research and development experts will assist you at every stage of your software project, from business needs analysis and ideation to product development and support.

Precise face and person attributes recognition
  • Face recognition at various pose angles and despite a shaky camera
  • Accurate face redaction
  • Hidden face detection and tracking
  • Person attributes detection: gender, age, headwear, and more

Object detection and tracking
  • Accurate object detection and tracking despite occlusions, poor lightning, etc.
  • Detection of any bag type, including abandoned objects and their owners
  • Custom object anonymization
  • Firefighter detection and tracking in a smoky environment
  • Unsupervised failure and anomaly detection

Accurate vehicle detection
  • Recognition of various vehicle types such as cars, bicycles, scooters, ships, and more
  • License plate recognition
  • Vehicle detection and tracking despite poor lighting and adverse weather conditions
  • Car parking monitoring

Streamlined M&E content processing
  • Accurate detection of ingested ads
  • Discovering anomalies in video content : black screens, glitches, artificial text, etc.
  • Scene analysis: detection of romantic, racing, disastrous, and other episodes
  • Spotting sensitive themes like smoking, alcohol, violence, and more

Accelerate development with our Content DNA know-hows

R&D Projects Productization

Intelligent ad placement
  • Accurately detecting ads of any type
  • Adapting commercials to particular scenes
  • Making advertisement less intrusive
  • Avoiding ad insertions in sensitive content

Deep tech

Sport highlights generation
  • Identifying any sports type
  • Detecting key moments in matches
  • Suggesting relevant content in real time
  • Automatically compiling engaging sport highlights

Connected world

Content personalization
  • Revealing hidden dependencies in scenes
  • Rapidly enriching content metadata
  • Going beyond traditional metadata like genres, actors, tags, etc.
  • Providing hyper-personalized content recommendations

Content DNA platform in action

Rapidly compiling soccer highlights for increased fan engagement
  • n-depth analysis of soccer videos, both live and VoD
  • Accurate identification of key moments in matches like goals, scoring chances, and fouls, as well as audience reactions
  • Rapid athlete-specific scene search powered by in-house face recognition SDK
  • Smart detection of scene replays
  • Customizable filters enabling manual highlight selection
  • Seamless highlights upload to multiple video hosting platforms
  • Intelligent ad placement

Face recognition SDK

We speeded up the development of a live TV solution, enabling our client to sharpen the competitive edge.
  • Multiple face detection in real time
  • Maximum accuracy and performance with any kind of input
  • Adjustable speed-to-accuracy parameters and trigger thresholds
  • Dynamic link library available for 32-bit and 64-bit versions of Windows and Linux
  • Ease of integration with C++/C#/Java code

Face recognition in action

Next-Generation WebRTC-Based Proctoring Platform.
  • Seamless WebRTC-based video stream capture and playback
  • Machine learning-supported face recognition based on a customized build of the OpenCV library
  • Automated session flagging based on a range of triggers
  • Scoring functionality that assigns penalty points for suspicious activity
  • User identity validation based on a biometric profile
  • Easy-to-use text and audio chat functionality
  • Advanced reporting and statistics across test-takers, exams, and educational establishments