Category: Java

Learning Invariant Color Features for Person Re-Identification

Robust Visual Tracking via Convolutional Networks without Training Abstract Deep networks have been successfully applied to visual tracking by learning a generic representation offline from numerous training images. However, the offline training is time-consuming and the learned generic representation may be less discriminative for tracking specific objects. In this paper, we present that, even without offline training with a large amount of auxiliary data, simple two-layerconvolutional networks can be powerful enough to […]


Robust Visual Tracking via Convolutional Networks without Training

Robust Visual Tracking via Convolutional Networks without Training Abstract Deep networks have been successfully applied to visual tracking by learning a generic representation offline from numerous training images. However, the offline training is time-consuming and the learned generic representation may be less discriminative for tracking specific objects. In this paper, we present that, even without offline training with a large amount of auxiliary data, simple two-layerconvolutional networks can be powerful enough to […]