Login / Signup
Higher-level representation of local spatio-temporal features for human action recognition using Subspace Matching Kernels.
Bisser Raytchev
Hideaki Kawamoto
Toru Tamaki
Kazufumi Kaneda
Published in:
ICPR (2016)
Keyphrases
</>
higher level
feature space
low level
representation scheme
feature representation
graph matching
fisher kernel
lower level
matching process
kernel methods
high dimensional data
low dimensional
image set
image matching
matching algorithm
keypoints
subspace learning
translation invariant
object recognition