Horus - The Invisible Made Audible
-
Upload
specktech -
Category
Technology
-
view
74 -
download
1
Transcript of Horus - The Invisible Made Audible
285 million people are estimated to be visually impaired worldwide:
39 million are blind and 246 have low vision*.
* Source: World Health Organization
Visually impaired285 M
Blind54 M
People with low vision (2015)
Blind39 M
Blind people (estimates for 2020)
Global data source: WHO, IBU
Horus is a wearable device developed to assist blind and visually impaired people during the entire day.
Horus observes, understands and describes the environment to the user, providing useful information.
Menu buttons
Stereocamera
Bone conduction system
Menu buttons
Volume buttons
Headset connection
Horus is composed of two parts: the wearable headset and the pocket unit
Our story (in short)
Feb 14’ 15’
SPECIAL AWARD
Recognized by MIT Tech. Review as a disruptive company
Jan 16’
GROWTH
12 new employment opportunities in two years
May 16’
Raised 900k$ in Venture Capital
BETA Prototypes
Received the first custom boards and assembled the first beta prototypes
Sep 16’
Project start
External input identification
Cameras acquire images
Images are transferred to the computing unit
3
2
1
Audio is transferred back to the headset
5Information extraction
4
Sound output
6
How Horus works
Horus can help the user with:
Scene descriptionFace recognitionText reading Object recognition Mobility assistance
Horus is able to recognize printed text. It helps the user in obtaining a correct position for reading, thanks to an audio feedback given through bone conduction.
Text recognition
If the text is located in the upper part of the field of view, Horus emits a high-
pitched sound to tell the user to lower the text
If the text is located in the lower part of the field of view, Horus emits a low-
pitched sound to tell the user to raise the text
Example of audio feedback
Horus can recognize people’s faces and describe them to the user. After detecting a face, Horus can quickly associate a name to it, thus learning and remembering the identity of the new person.
Face recognition
Horus is able to recognize an object thanks to 3D perception.
When Horus has to learn a new object, the user can simply rotate it in front of the cameras. Horus will memorize and learn its shape in order to properly recognize it in different situations
Object recognition
Processing
Input frame (stereo)
Point cloud
Point cloud
Disparity map
Processing pipeline
Center sound
Right sound
High pitch sound
Low pitch sound
Left sound
From above
The location of the object is communicated using stereo audio and variation in pitch
Audible UI
Horus can observe and analyze the surrounding environment to detect the presence of potential obstacles. The user is made aware of their presence through a 3D sound map.
Mobility assistance
User
High-pitched sound left
Horus uses 3D sound to report the presence of obstacles during movement. The space in front of the user is divided in different sectors: lateral obstacles generate high-pitched sounds in one of the two speakers, while
central obstacles generate low pitched centered sounds. These sounds are repetitive and they increase in repetition frequency as the obstacle gets closer.
High-pitched sound right
Reporting obstacles
Low-pitched sound center
Horus is able to describe what the cameras are seeing.Whether it is a postcard, a photograph or a landscape, the device provides a description of what is in front of it with a sentence.
Scene and photos description
Video Here
What did we learn?
1
Speaking a lot about our idea
2
Asking for help to mentors and advisors
3
Write everything down
4
Hiring is super important!
5
Always have a plan B(and maybe C, D, and so on…)
it.linkedin.com/in/lnardelli
www.horus.tech
Luca Nardelli - CTO
Interested in working with us?Visit www.eyra.io