Using AI technology to help the visually impaired regain their sight is pretty cool!
[Copy link]
Artificial intelligence (AI) developer Jagadish K. Mahendran and his team have designed an AI-powered, voice-activated backpack that can help the visually impaired navigate and perceive the road. The backpack can help detect common road obstacles such as traffic lights, hanging barriers, crosswalks, moving objects, and uphill and downhill slopes, all running on a low-power interactive device.
“Last year, I met a friend who has visual impairments, and I realized the irony was that while I had been training robots to see, there were a lot of people who couldn’t see and needed help. This inspired me to build this visual assistance system, using Intel technology’s OpenCV AI Toolkit (OAK-D).”
The World Health Organization estimates that there are 285 million visually impaired people worldwide. At the same time, visual assistance systems to help them get around are currently limited, ranging from voice-assisted smartphone applications based on the Global Positioning System (GPS) to smart canes equipped with cameras. These devices and systems available to the visually impaired are not only very limited, but also lack depth perception capabilities, which is very necessary for the visually impaired to travel independently.
“It’s incredible to see how quickly this developer has developed a solution using Intel’s AI technology at the edge to help these visually impaired people live better,” said Hema Chamraj, director of technology advocacy and AI4Good at Intel. “Now that we have the technology, we need the imagination of the developer community to take it further.”
Working principle:
The system is housed in a small backpack that contains a host computing device, like a laptop. The camera is hidden in the vest jacket, and a pocket-sized battery pack is placed in a waist pack, which provides about eight hours of power. A Luxonis OAK-D Spatial AI camera can be attached to the vest or waist pack and then connected to the computing device in the backpack. There are three small holes on the vest for the OAK-D camera, and the camera is located on the inside of the vest.
Backpacking device
“Luxonis’ mission is to enable engineers to build things that matter to people, and to help them quickly leverage Intel’s AI technology,” said Brandon Gilles, founder and CEO of Luxonis. “It’s exciting to see someone develop something as remarkable as the AI-powered backpack using OAK-D in such a short time.”
Wearable Devices
The OAK-D device is a powerful, all-in-one AI device that uses the Intel Movidius VPU and Intel OpenVINO toolkit for on-chip edge AI inference. It is capable of running advanced neural networks while providing accelerated computer vision capabilities and real-time depth maps from its stereo pair and color information from a 4K camera.
Using a Bluetooth headset, users can use the system through voice commands, and the system will respond with voice commands. As users move around, the system will provide voice information about common obstacles, including traffic lights, tree branches and pedestrians. It can also warn of intersections, curbs, stairs and entrances ahead.
I hope that this device can be mass-produced and put on the market as soon as possible, using AI technology to help more visually impaired people regain their sight.
|