New AI Sees Around Corners Using Fish-Eye Cameras
This could make self-driving cars and security cameras safer and cheaper.
This article is about arXivLabs, a framework that lets collaborators develop and share new arXiv features directly on the website. Both individuals and organizations working with arXivLabs have embraced its values of openness, community, excellence, and user data privacy — and arXiv says it only works with partners who adhere to those values. If you have an idea for a project that would add value for arXiv's community, you can learn more about arXivLabs.
Note: the source provided contains no information about FUSEye, fish-eye cameras, object detection, training data, surveillance, or autonomous vehicles, so none of that can be included.
- Fish-eye cameras see more but distort images, which normally confuses AI.
- FUSEye uses overlapping views and zero-initialized adapters to fix distortion with less training data.
- This could lead to cheaper, more effective surveillance and self-driving car systems.
Why It Matters
Wider, cheaper AI vision could make self-driving cars safer and security cameras more effective.