81
Voted
ICIP
2009
IEEE
14 years 7 months ago
2009
IEEE
We present a new method of computing invariants in videos captured from different views to achieve view-invariant action recognition. To avoid the constraints of collinearity or c...