CrowdAI News

Running a computer vision model on a cell phone or mobile device is a powerful tool that can enable real-time analysis of images and videos, which can be useful in a variety of applications. While there are challenges to streaming computer vision models on small devices, CrowdAI has developed a roadmap of techniques and tools to overcome these challenges. By leveraging cloud driven API connections for invoking inference from a trained model, CrowdAI sees a pathway to real-time analysis of imagery and video on small devices operating at the edge. Additionally, the geospatial benefits of building models from media captured on cell phones can offer unique advantages for training, monitoring, and analyzing objects of interest.

In today's world, where we are surrounded by computers and cameras of all types and sizes, it's essential for machine learning services to be deployment-agnostic and camera-agnostic. Being able to work in any cloud, hardware, or software environment; and to use any camera or sensor is an invaluable advantage that has become increasingly important in recent years as the use of cameras has exploded in various industries. These features allow for greater flexibility and ease of use—exactly what CrowdAI strives to provide—enabling ML to be used in a wider range of applications.

Going from a complex segmentation model to a simpler bounding box object detection model using SAM may seem like a bit of overkill, but there are some instances where an object detection model is favored over a segmentation model. For example, if we have a photo of a street with a bunch of pedestrians, a detection model can provide insight into how many people are there, their location in the frame, and how they interact with each other; segmentation masks wouldn’t give us as useful information since they would just be silhouettes of standing or walking people. Another benefit is that object detection models are designed to be more robust to variations in object size, rotation, and aspect ratio, making them ideal for identifying objects with diverse geometries. Lastly, when computational resources are limited, object detection models tend to be less computationally intensive than segmentation models, which can require more processing power and memory to run efficiently. 

To test out quickly moving from an object detection model to segmentation model using the SAM integration in the CrowdAI platform, we revisited a detection model we had built in the past. CrowdAI worked on automating visual inspection for a real estate developer. Their goal was to have tenants take photos of damage in their apartments and have our models automatically identify what the potential problem could be and on what appliance. The examples below are from a quick study CrowdAI completed to identify if a shower head was leaking – for the purpose of the study we used images of shower heads running and not running to train a model to identify water ‘leaking’ from the shower head. 

Segmentation is a core computer vision architecture - training a model based on the exact pixels which make up an object of interest. This computer vision architecture typically requires a time consuming annotation process because a large number of carefully labeled media are needed to train a model, with the labeling work often needing to be done by subject matter experts. The caveat being, most subject matter experts don’t have the time to be meticulously labeling, nor do they have hundreds of examples of each object, defect or anomaly they are aiming to detect. This creates a huge roadblock for producing quick, off the shelf or domain specific customized segmentation models. 

Data-centric AI is all the buzz. In this post, CrowdAI looks at a few examples to explore advantages and limits of using pre-trained models for automated labeling. 

Finding a defective item amongst thousands of non-defective objects is an essential task across industries. In this post, the CrowdAI research team takes a deep dive into how anomaly detection works using the k-NN model.

The term "big data" gets thrown around a lot, but what does it mean in the context of deep learning?

When we’re not working on hurricane or wildfire models—or, I guess, curling (full disclosure, contrary to what the job description claims, I've yet to go curling)—CrowdAI builds segmentation models to try to pick out static objects in full-motion video.

Our goal at CrowdAI is to make artificial intelligence more accessible, and providing model outputs in an accessible manner closes the gap between no-code AI solutions and delivering results for business analysts. Using ChatGPT and GPT-3 in conjunction with CrowdAI helps business analysts focus on deriving insights instead of interpreting the results.

CrowdAI is open-sourcing our Hurricane Ida data for anyone to use under a Creative Commons license.

AI’s next evolution for government, enablement tools, will attempt to overcome previous limitations while also presenting new challenges.

Hurricane Laura was one of the most powerful storms to strike the U.S. in quite a long time. As we've done for natural disasters in the past, CrowdAI used our custom-built building damage model on NOAA's Emergency Response Imagery to create a dataset to help assess the damage from the storm.

At CrowdAI, we help people make the most of their imagery—whatever the source. One type of imagery we’ve seen increase in popularity in recent years is infrared (IR) or thermal imagery.

Tornadoes are unpredictable. Unlike other storms, where we may have a few days to prepare and evacuate, a tornado's exact point of formation and its path are very difficult—if not impossible—to predict.

The aftermath of a hurricane is a disorienting time for families and businesses alike. Given the strong links between rising sea surface temperatures in the Atlantic and hurricane frequency and power.

Hurricane Dorian formed on August 24, 2019, wreaking havoc on hundreds of thousands of people—from the Carribean all the way to Canada.

The extent of the damage from Camp Fire in Northern California is unimaginable. It's difficult to wrap your head around it without walking around each affected neighborhood.

With the incredible uptick in oil and gas exploration in recent years, the U.S. has seen a lot of changes to the landscape in regions across the country.

Global Airport Monitoring at Your Fingertips: How Few-Shot Learning Is Becoming Table Stakes for Automated Target Recognition

CrowdAI, the SanFrancisco-based computer vision company that helps organizations automate visual inspection, was selected among hundreds of candidates as one of the World Economic Forum’s “Technology Pioneers”.

CrowdAI today announces a partnership with i3ICS of Fairfax, Virginia, a service-disabled veteran owned small business(SDVOSB). Through this partnership, i3 ICS offers CrowdAI computer vision technologies and professional services under its multiple schedule award (MSA)with the General Services Administration (GSA).

CrowdAI Public Sector welcomes Keith E. Johnson as Strategic Revenue Officer and Melanie Corcoran as Federal Advisor. 

CrowdAI, the leading no-code computer visionplatform, today announced a strategic go-to-market alliance with Carahsoft Technology Corp., the Trusted Government IT SolutionsProvider®.

CrowdAI, the leading no-code computer vision platform, today announced it signed a five-year Cooperative Research and Development Agreement (CRADA) with the Space Vehicles Directorate of the Air Force Research Laboratory (AFRL). The agreement gives AFRL access to CrowdAI’s cutting-edge deep-learning tools and experts, while enabling personnel from both organizations to experiment with the country’s hybrid space Intelligence, Surveillance, and Reconnaissance (ISR) architecture and computer vision. 

CrowdAI accounces its selection as an institutional AI partner under a $249 million-ceiling Blanket Purchase Agreement issued by the Department of Defense (DoD) Joint Artificial Intelligence Center (JAIC).

CrowdAI is proud to announce an AFWERX Small Business Innovation and Research (SBIR) Direct-to-Phase II award from the U.S. Air Force (AF). The project will equip U.S. Southern Command (SOUTHCOM) with AI capabilities that enable automated, near-real-time situational awareness for humanitarian assistance and disaster response (HADR), as well as support countering illicit trafficking in Latin America and the Caribbean.

CrowdAI announced today that it won an AFWERX Small Business Innovation and Research (SBIR) Phase II award from the U.S. Air Force (AF). The project will explore CrowdAI’s artificial intelligence (AI) capabilities to improve building maintenance at AF bases across the country.

CrowdAI announced today that it won an AFWERX Small Business Innovation and Research (SBIR) Phase I award from the US Air Force (AF) and the National Geospatial-Intelligence Agency (NGA).

At CrowdAI, our goal is to empower you to create your own computer vision (CV) models to simplify your work.

In the real world, datasets are almost never straightforward. Deep learning offers a way to more quickly and accurately analyze visual datasets.

What exactly are neural networks, and why are they so important to deep learning?

Deep learning has the potential to change how everyone interacts with AI. But why? This is the first in our article series on the power of deep learning.

Following a few simple best practices will help you collect the imagery you need to make your AI efforts successful. Here's our cheat sheet!

Our tips on how to get your AI projects out of R&D: involve the right Subject Matter Experts (SMEs) early and often throughout the project.

There's a better way to understand and consider metrics/KPIs when it comes to AI model development. Here's our recommendation for getting out of R&D.

One of the main reasons projects get stuck in R&D is that projects are often scoped only to the data you have on hand. Here's how to help fix that.

CrowdAI has recommendations for how to get your computer vision AI project out of research and development (R&D).

What happens if you don’t have imagery and video but already see the value from automating certain manual, visual tasks? Here's how to get started.