Learning to traverse image manifolds
Publikation: Bidrag til tidsskrift › Konferenceartikel › Forskning › fagfællebedømt
We present a new algorithm, Locally Smooth Manifold Learning (LSML), that learns a warping function from a point on an manifold to its neighbors. Important characteristics of LSML include the ability to recover the structure of the manifold in sparsely populated regions and beyond the support of the provided data. Applications of our proposed technique include embedding with a natural out-of-sample extension and tasks such as tangent distance estimation, frame rate up-conversion, video compression and motion transfer.
Originalsprog | Engelsk |
---|---|
Tidsskrift | Advances in Neural Information Processing Systems |
Sider (fra-til) | 361-368 |
Antal sider | 8 |
ISSN | 1049-5258 |
Status | Udgivet - 2007 |
Eksternt udgivet | Ja |
Begivenhed | 20th Annual Conference on Neural Information Processing Systems, NIPS 2006 - Vancouver, BC, Canada Varighed: 4 dec. 2006 → 7 dec. 2006 |
Konference
Konference | 20th Annual Conference on Neural Information Processing Systems, NIPS 2006 |
---|---|
Land | Canada |
By | Vancouver, BC |
Periode | 04/12/2006 → 07/12/2006 |
ID: 302051790