Crater Labs Inc.
$75,000.00 CAD
- Department
- National Research Council Canada
- Recipient country
- Canada
- Fiscal year
- 2018-2019
- Agreement period
- January 7, 2019 – April 12, 2019
- Reference
- nrc-cnrc:172-2018-2019-Q4-921943
Published purpose
We are seeking to develop a flow-based computer vision algorithm to support our ‘electronic guide dog’. This algorithm will allow our system to not only recognize objects, but determine the rate of speed at which objects are approaching, and further determine what kind of obstacle is upcoming. For example, without flow, computer vision systems will recognize an escalator as stairs, and a moving walkway as a floor. With a flow-based computer vision algorithm, we will be able to distinguish between stairs and escalators, and provide feedback to help visually impaired people gauge the level of caution required. This will improve the safety of the visually impaired, detecting obstacles a guide dog cannot.
Community tags
Tags are applied by readers, not by this site. One tag per person per record.