Services — 

Computer Vision for Supply Chain

Make faster, more accurate decisions on the ground with computer vision across your supply chain.

Get a demo -->
a woman with short brown hair and man with blond hair and glasses, viewing warehouse insights on a desktop with pallets and measurements graphics along with colorful artificial intelligence graphics overlaid

We've worked with some of the world's largest brands.

shield icon
shield icon
shield icon
shield icon
shield icon
shield icon

What does computer vision capability look like?

man driving a forklift, with a pallet of boxes, driving it into a truck, with navigation and tracking graphics overlaid

Taking advantage of edge computing, we can give you the capability to provide real insights into what is actually happening in your warehouse or terminal.

We can provide the flexibility to model out a wide-variety of scenarios to get you the information you need.

Let's look at some use cases

Hard Hat Detection

  • Detects employee compliance for wearing PPE (such as hard hats)
  • Predict correlations between time of day and increased violations

With computer vision, you can run Machine Learning (ML) inference on video feeds to detect hazardous conditions in warehouses or distribution centers. For example, if hard hats are required for all forklift drivers, we can build a solution that detects when a driver is not wearing the required hard hat, send an alert immediately on-site, AND send data of the violation to your existing ERP system for later analysis.

man with short hair and beard holding a tablet checking in on a warehouse report

Text & Barcode Scanning

  • AI model provisioned to run at key points in the warehouse process (on forklift for example)
  • Product information is sent to the operational system (WMS) for confirmation and instructions are relayed to the associate or tool.

Making sure that you are identifying the right product can dramatically accelerate your automation journey. Our computer vision capabilities can help you and your team make sure the right product is being picked through text and barcode scanning. Our model can identify that a package has a barcode and check that information against WMS tools to ensure the right action is taken on that tool.

Product Counting & Tracking

  • Detects number of products in a specific bin or area and checks for accuracy
  • Upload flagged events to Cloud Storage for further analytics and actionable insights

Tracking stock and controlling the flow has never been more complex – and important to figure out. Unfortunately, human processes are just not cutting it. Computer vision can capture images of products and track the movement via barcodes, which results in less lost or mis-identified inventory. Whether you need to track product on a production line, going in and out of warehouses, or coming into your retail store, computer vision can nearly eliminate product tracking errors.(Psst…see how we solved this $60M problem for a manufacturing client here.)

young woman with long dark hair holding a tablet, checking a warehouse, pallet measurement alert

Pallet Measurement

  • AI model provisioned at the warehouse’s edge at key junctions or traffic areas.
  • Model identifies that a pallet has entered the area and takes measurements.

Most measurements and tracking today are handled by frontline employees. These tasks are mundane, overly manual, and error prone. Through computer vision, you can measure pallets as they arrive and track those pallets from storage to shipping. With more accurate measurements and locations, you’ll have less wasted space in your DCs and trucks, and more accurate reporting to customers.

What is the Google Cloud Edge Appliance?

    Running heavy data processing in the cloud isn’t possible or attainable, often due to a lack of network connection or high cost.  In addition, only relevant data needs to be sent to the cloud to reduce cost of ingress and egress. Existing edge devices are usually limited in their computing power and teams lack visibility and management. 

    Edge computing simplifies data collection, analytics, and processing at far-edge locations. Your data is stored on the appliance, where containerized applications process the data locally using ML inference, aggregation, and custom logic to generate insights. Configurable transfer jobs easily move aggregated insights and relevant data to Cloud Storage when connectivity and bandwidth are available.

    Learn more about our Google partnership here.

    Computer Vision Proof of Concept

    90-day proof of concept to validate the business value of your solution.

    Business validation

    Use case & success definition

    Working initial model

    Prototype of scaled solution

    Ready to get started? We’ll kick things off by creating a proof of concept for your computer vision solution. This engagement will validate the need, prove ROI, and show what’s possible with a working prototype of the computer vision solution.

    Get started ->

    Local Data Processing

    AI, analytics, and database solutions uncover insights and remove traditional constraints of scale, performance, and cost.

    Scalability through MLOps

    Gain benefits of high-powered computing across devices in multiple locations.

    • Improve and test models through enterprise grade platforms like GCP and VisionAI
    • We’ll work with your team from everything from annotation, logging,  model registry and production deployments
    • PyTorch or Tensorflow support for your team’s desired tech stack needs
    • State of the Art (SOTA) model and MLOps knowledge to keep you ahead 
    • Allow your machine learning engineers to focus on model creation and value while we handle operational logistics

    High-Powered Edge Computing

    Run high-power functions at the edge to enable computer vision and Google AI to get real-time insights on location.

    Centralized Management

    Cloud-backed control plane powered by Anthos that provides a consistent management experience at scale.

    [Case Study]

    $60M is no small change

    One of the world’s leading food processors was losing millions of dollars annually because of an overly manual and error-prone process that resulted in a higher-than-tolerable instance of misidentified and lost inventory on the factory floor. Leaders knew the existing process was woefully inefficient and riddled with repeated patterns of human error. Nearly all agreed that significant change was needed. Namely, a much more efficient system that reduced the load on its factory staff while increasing the accuracy and reliability of the labeling process.

    Mobile phone with financial report application

    We've got the technical chops.

    Computer Vision

    Data Modeling

    Data Wrangling

    ML Ops

    Vision AI FAQs 

    What is computer vision?

    Computer vision focuses on creating digital systems that can process, analyze, and make sense of visual data similar to humans.

    What is an edge appliance?

    An edge appliance is a secure, high-performance appliance for edge locations. It provides local storage, ML inference, data transformation, and export.

    Who should use an edge appliance?

    Edge appliances are ideal for use cases where bandwidth and latency limitations prevent organizations from processing the data from devices like cameras and sensors back in cloud data centers. These appliances simplify data collection, analytics, and processing at remote locations where copious amounts of data coming from these devices need to be processed quickly and stored securely.

    Enterprises from the supply chain, manufacturing, and automotive verticals with low-latency and high throughput requirements can benefit from edge computing.

    What is the technical setup of the Edge Device look like? How does it fit into my existing infrastructure?

    We build it for your unique environment. Using existing camera systems, developing software that will plug seamlessly into your ERP, WMS, TMS, or existing platforms. It’s important to note, there is not a one-size fits all solution, so we will tailor to your exact needs.

    What are some of the challenges of Computer Vision at scale?

    • Mass amounts of data at their locations from cameras, sensors, and other devices
    • Data needs to be quickly processed to generate actionable insights on location
    • Running heavy data processing in the cloud isn’t possible (maybe due to a lack of network connection or high cost)
    • Only relevant data needs to be sent to the cloud to reduce cost of upload / download
    • Limited computing power on existing edge devices
    • Lack of visibility and management with existing on-site appliances

    Seamless scaling with
    internal teams

    shield icon

    Faster path to value from implementation to business outcome

    shield icon

    Allow customer’s internal teams to focus on other work

    shield icon

    Seamless transition back to internal teams after implementation

    We know how to jump-start your client’s Google Edge Device implementation and bridge the gap between business and technology.Whether your client needs help accelerating their internal effort, or have the whole project handled externally, our team can get it done. Faster.

    diverse group of office team members working together looking at a laptop, conversing, with cloud graphic elements

    Request a consultation with a computer vision expert.

    Let’s talk about what computer vision can do for your business. We’ll walk you through some of the work we’ve done and figure out where you could get the most value from a computer vision solution.

    Thank you! Someone from our team will be in touch shortly.
    Oops! Something went wrong while submitting the form.

    Creating a More Outcome-Driven Organization

    POC young female with glasses in  in discussion with another team member

    Three key focus areas every leader should be thinking about right now in order to create a more collaborative, outcome-focused culture

    Dive in -->

    How digital twin can change the face of retail (It’s easier than you think)

    a male and female retail worker logging in tasks into a tablet with a cloud and dashboard graphics

    Digital twin is not just a buzzword; it holds a lot of power and opportunity for retail organizations to improve their operations and get real-time information about the customer in ways that other data analytic strategies can't.

    Learn more -->
    Case Study

    Using machine learning to solve a $60M inventory loss problem

    group of workers pulling and packaging food into crates

    A closer look at how the emerging tech and analytics teams applied machine learning to Operational Technology, increasing product identification accuracy to over 90%

    Explore Our Work -->