Network in network based weakly supervised learning for visual tracking
We propose a novel tracking method using a deep network, i.e. Network in Network (NIN).A method to transfer features learned from NIN on the source tasks to the tracking tasks.The samples are simultaneously utilized to address the drifting problem. One ...
Adaptive visual target detection and tracking using weakly supervised incremental appearance learning and RGM-PHD tracker
Display Omitted A multiple visual target detection and tracking system is proposed.Relative keypoint-based appearance model is proposed for visual targets.Incremental learning method with weak supervision is introduced to update target appearances.The ...
APPOS
Our occlusion segmentation method can be adapted to most kinds of tracking methods.The segmentation process will not affect the tracking results mostly.The candidate region evaluating process can get back the tracking-lost vehicle. In traffic ...
Local consistent hierarchical Hough Match for image re-ranking
Re-ranking algorithm special for small vocabulary.Hierarchical Hough Voting from local to global.Being able to store in mobile visual search systems. Geometric image re-ranking is a widely adopted phrase to refine the large-scale image retrieval systems ...
3D object retrieval based on sparse coding in weak supervision
The proposed method does not require the explicit virtual model information.We utilized FDDL to learn dictionary to reconstruct query sample for retrieval.Our approach has high generalization ability and can be used on other applications. With the rapid ...
Towards robust subspace recovery via sparsity-constrained latent low-rank representation
We present a Sparse Latent Low-rank representation approach for robust visual recovery.This approach constructs the dictionary using both observed and hidden data.A low-rank representation with enhanced sparsity can be derived.Extensive experiments have ...
An aerial image recognition framework using discrimination and redundancy quality measure
Aerial image categorization framework.Discriminative subgraphs.Subgraph mining.SVM. Aerial image categorization plays an indispensable role in remote sensing and artificial intelligence. In this paper, we propose a new aerial image categorization ...
An improved augmented reality system based on AndAR
We present an adaptive Augmented Reality (AR) architecture.The improved AndAR can augment reality based on natural features.The solution can augment the reality anywhere.The presented approach improves system speed and extends tracking range. AndAR is a ...
Genetic algorithm and mathematical morphology based binarization method for strip steel defect image with non-uniform illumination
We proposed EOBMM to reduce the non-uniform illumination in strip steel defect image.We put forward BMBGA to decide the threshold value of defect image binarization.We combined EOBMM and BMBGA to present strip steel defect image binarization method.The ...
