Built utilizing YOLOv2 model, Cloudtect is able to effectively identify and track specified objects in real-time. Through machine learning and regression, this software is able to identify various objects within various frames . This process of using cloud computing in conjunction with OpenCV and TensorFlow has not been achieved in this capacity before.
With this computer vision-based application, the goal is to grow abilities in a variety of applications:
This means that anyone has the capability to develop and learn machine learning object detection without installing libraries and the need to have excessive computing power. This is because the object detection program is running on an external software.
Cloudtect Object Detection
YOLOv3 Image Detection model used to classify various objects in real-time.
Through object detection being used in conjunction with cloud computing allows for a larger amount of computing power and for algorithms and methods to run much quicker.
Instead of needing expensive hardware and large teams to test programs for object detection, this method will allow for the process to be exponentiated, and allow for new discoveries to be uncovered.
Allowing for cloud-computing object detection in real-time can drive research that can change the world.
This technology will allow anyone to utilize machine learning and object detection methods.
A large reason why object detection is being limited in usage is the lack of training for employees and interns due to storage capacity, time constraints, and computing power.
With Cloudtect, the utilization of cloud computing can allow for real-time experimentation and implementation of algorithms and object detection. Without the need to have all students download and train machine-learning methods, it allows for seamless integration of development into curriculums.
Cloudtect allows a way for students within a classroom or lab environment to utilize object detection and machine learning methods without requiring a large amount of personal computing power or storage capabilities.
Similar to the training functionality, by utilizing external servers, it can drive learning to be much more effective and efficient.
With machine learning becoming more relevant everyday, there needs to be an efficient way to teach students
An example of object detection completing tracking around a store for hot spots.
By having an external server doing mass calculations for algorithm-based programs, retail stores can easily utilize the tracking for analytics.
Looking at customer positions around the store, managers can recognize the hot spots of the store and drive sales through data being processed through object detection.
We are a group of friends that wanted to use QHacks to make an impact on machine learning and cloud computing.
Hi there! I am a second-year Computer Engineering Student at Queen's who has a huge passion for technology and business! I'd love to hear what you think about our product!
I am a second year Applied Mathematics Engineering student. I am driven to create software that will improve people's day to day lives. My areas of interest are creating software for education and medicine.
I am a second year Applied Mathematics Engineering student, Computer Option. I have a lot of interest in machine learning implementations, data science, and shawarma.