We are happy to share the excitement from the article which NVIDIA, well-known GPU’s provider and AI focused tech company, published about DataFromSky – our AI in the Sky!
As the article further explains, we train our deep neural networks using thousands of hours of video footage from around the globe, shot in various weather conditions. The training takes place on NVIDIA GPUs using Caffe and TensorFlow. These specialized neural networks can then recognize objects of interest and continually track them in video footage. The data captured via this AI process is used in numerous research projects, enabling deeper analysis of object interaction and new behavioral models of drivers in specific traffic situations. Ultimately, this kind of data will also be crucial for the development of autonomous vehicles!
We are also carrying out extensive optimizations of the whole architecture, moving to NVIDIA Jetson AGX Xavier, in order to be able to analyze data in real time. Our current work focus is also on the development of an online monitoring drone, which will be able to evaluate image data in real-time, directly during its flight. This type of drone can be used, for example, as a substitute for a police helicopter, or can operate as a mobile surveillance system for personal and property protection.
Read the full article on NVIDIA blog at this link.
We are also a member of the NVIDIA Inception program – virtual accelerator program led by NVIDIA nurtures dedicated and exceptional technological companies, who are revolutionizing industries with advances in AI and data science and help them in all stages of product development, prototyping, and deployment.
DataFromSky is a unique solution for motion analysis of aerial video data. The ability to detect, track and classify moving objects from aerial videos provided by drones, fixed cameras or other video sources opens new possibilities in the field of traffic monitoring. The technology behind is under intensive development and grows more efficient every day through the embedded capability of learning from its mistakes. We are happy to announce that we have made a significant step towards enhanced accuracy thanks to the complex integration of deep neural networks into DataFromSky framework. Our deep neural networks are being trained on millions of samples 24 hours a day. The quality is quite impressive right now!
Our tool for trajectory interpretation DataFromSky Viewer also comes with a new functionality. The new version extracts parameters which are sometimes difficult to obtain, such as gap-time, time to follow, heatmaps, and dimensions of vehicles. We recognize the potential of the data we are able to obtain, and, therefore, co-work on our tools with traffic researchers and traffic engineers worldwide. The video below demonstrates the new functionality of DataFromSky Viewer. Take a look!
A Workshop titled Modern methods of image recognition and processing methods was organized recently by Technical University in Liberec. The aim of the workshop was to offer insight into image processing applications from the industry, as opposed to research setting.
We presented principles of detection and classification of vehicles from aerial video data using Deep Neural Networks. As you can imagine, Deep Neural Networks are integral part of DataFromSky platform and are used in many subtasks of DataFromSky solution. If you want to know more about the magic behind, you can download the presentation (PDF, 2 MB) and/or contact us.