Retail

Machine vision can significantly enhance the retail sector by streamlining operations, improving customer experience, and boosting overall efficiency. In inventory management, machine vision systems can automatically track stock levels, identify misplaced items, and reduce shrinkage by detecting theft or loss in real time. For customer experience, these systems can analyze shopper behavior, helping retailers optimize store layouts and product placements to enhance the shopping journey. Moreover, machine vision can speed up checkout processes through automated product recognition, reducing the reliance on barcodes and minimizing human error. This technology also enables personalized marketing by analyzing demographic information and purchasing patterns, allowing retailers to tailor promotions and advertisements more effectively. Additionally, machine vision can improve customer experience by enabling features like virtual try-ons or interactive displays that respond to customer interactions, making shopping more engaging and personalized. Overall, machine vision brings a high level of automation and intelligence to retail, driving operational excellence and significantly enhancing customer satisfaction.

Virtual wardrobe application - Software & Algorithms

The Need

The client is a startup company that implemented a fashion social network. One of the important services that the network provides to the users is a virtual wardrobe. Building the personal wardrobe is done by each user uploading pictures with their smartphone of clothes they own or just bought. The user can take free-hand pictures with unconstrained background and with no additional data attached to the uploads. 

The Challenge

  • Separating fashion items from the background that changes from image to image.
  • Enhancing the images so they look good on the web.
  • Performing all these actions in a short time to not affect the user experience.

virtualWardrobe

The Solution

Design and development of an integrated software module that applies segmentation of fashion items from the background. After the segmentation, the software categorizes the fashion items, determines their colors and processes the images to be graphically attractive before adding them to the virtual wardrobe of the user. Finally, a comparison with known brands is made using AI so more commercial data on the uploaded items can be displayed and shared. In order not to disrupt the user experience, the software and algorithm takes less than a second and the process is guided with an intuitive HMI.

3D analysis of faces - Software & Algorithms

The Need

The client is a startup company that creates a new costumer experience while purchasing eyewear. The new experience is based on a preliminary virtual demonstration of the eyeglasses before purchase. The customer can choose size, model, color and lens and virtually see them all together on their face before the actual manufacturing.


The Challenge

  • Real time 3D reconstruction.
  • Complex points cloud analysis of various faces.

3d anlysis of faces glasses

The Solution

Design and development of an integrated software module that reconstructs a precise 3D model of the costumer’s face from multiple images. From that model, crucial facial features located on the nose, ears and eyes are extracted in real time to finally fit and display the eyeglasses on the customer’s virtual face.

Analyzing and counting people – Software & Algorithms

The Need

The client is a startup company developing tools for counting people in public places for further commercial analysis. One of the more complex places that counting needs to be done is in a theatre. To do that, the counter needs to be light invariant, accurate, dynamic and maintain the privacy of the un-interfered audience.


The Challenge

  • Handling extreme and variable lighting conditions.
  • Dealing with unexpected movements of people in the space.

The Solution

Design and development of an integrated software module that can count people and detect and follow their movements in the theatre during the show. The system uses special wavelengths to work freely in every light condition without interfering the audience and without violating its privacy. Special blob analysis and video motion detection algorithms were implemented to handle the split and merge problem of people in the theatre and to get an accurate and continuous counting.

Solutions for a variety of industries

Join Our Newsletter
News & Articles

24 AI Companies in Israel

As Israel is categorically a global hub for Artificial Intelligence (AI), it boasts a large group of service providers offering consulting and custom development of

Read More »