In this work, we present a novel segmentation method for deformable objects in monocular videos. Firstly we introduce the dynamic shape to represent the prior knowledge about obje...
State of the art methods for image and object re-
trieval exploit both appearance (via visual words) and
local geometry (spatial extent, relative pose). In large
scale problems,...
Michal Perdoch (Czech Technical University), Ondre...
A novel video object segmentation method is proposed which aims at combining color and motion information. The model has a multilayer structure: Each feature has its own layer, cal...
We demonstrate a novel approach to modelling arbitrary temporally-deforming objects using spatio-temporal Fourier descriptors. This is a continuous boundary descriptor, which can ...
Real world images often contain similar objects but with different rotations, noise, or other visual alterations. Vision systems should be able to recognize objects regardless of ...