A coupled discriminative dictionary and transformation learning approach with applications to cross domain matching

Cross domain and cross-modal matching has many applications in the field of computer vision and pattern recognition. A few examples are heterogeneous face recognition, cross view action recognition, etc. This is a very challenging task since the data in two domains can differ significantly. In this work, we propose a coupled dictionary and transformation learning approach that models the relationship between the data in both domains. The approach learns a pair of transformation matrices that map the data in the two domains in such a manner that they share common sparse representations with respect to their own dictionaries in the transformed space. The dictionaries for the two domains are learnt in a coupled manner with an additional discriminative term to ensure improved recognition performance. The dictionaries and the transformation matrices are jointly updated in an iterative manner. The applicability of the proposed approach is illustrated by evaluating its performance on different challenging tasks: face recognition across pose, illumination and resolution, heterogeneous face recognition and cross view action recognition. Extensive experiments on five datasets namely, CMU-PIE, Multi-PIE, ChokePoint, HFB and IXMAS datasets and comparisons with several state-of-the-art approaches show the effectiveness of the proposed approach. (C) 2015 Elsevier B.V. All rights reserved.

Stimuli-responsive protamine-based biodegradable nanocapsules for enhanced bioavailability and intracellular delivery of anticancer agents

Enzyme-and pH-responsive polyelectrolyte nanocapsules having diameters in the range of 200 +/- 20 nm were fabricated by means of Layer-by-Layer assembly of biopolymers, protamine, and heparin, and then loaded with anticancer drug doxorubicin. The incorporation of the FDA-approved peptide drug protamine as a wall component rendered the capsules responsive to enzyme stimuli. The stimuli-responsive drug release from these nanocapsules was evaluated, and further modulation of capsule permeability to avoid premature release was demonstrated by crosslinking the wall components. The interaction of the nanocapsules with cancer cells was studied using MCF-7 breast cancer cells. These capsules were readily internalized and disintegrated inside the cells, culminating in the release of the loaded doxorubicin and subsequent cell death as observed by confocal microscopy and MTT Assay. The bioavailability studies performed using BALB/c mice revealed that the encapsulated doxorubicin exhibited enhanced bioavailability compared to free doxorubicin. Our results indicate that this stimuli-responsive system fabricated from clinically used FDA-approved molecules and exhibiting minimal premature release has great potential for drug-delivery applications.

Anomalous reduction in domain wall displacement at the morphotropic phase boundary of the piezoelectric alloy system PbTiO3-BiScO3

A comparative study of field-induced domain switching and lattice strain was carried out by in situ electric-field-dependent high-energy synchrotron x-ray diffraction on a morphotropic phase boundary (MPB) and a near-MPB rhombohedral/pseudomonoclinic composition of a high-performance piezoelectric alloy (1-x) PbTiO3-(x)BiScO3. It is demonstrated that the MPB composition showing large d(33) similar to 425 pC/N exhibits significantly reduced propensity of field-induced domain switching as compared to the non-MPB rhombohedral composition (d(33) similar to 260 pC/N). These experimental observations contradict the basic premise of the martensitic-theory-based explanation which emphasizes on enhanced domain wall motion as the primary factor for the anomalous piezoelectric response in MPB piezoelectrics. Our results favor field-induced structural transformation to be the primary mechanism contributing to the large piezoresponse of the critical MPB composition of this system.

CROWD FLOW SEGMENTATION IN COMPRESSED DOMAIN USING CRF

Crowd flow segmentation is an important step in many video surveillance tasks. In this work, we propose an algorithm for segmenting flows in H.264 compressed videos in a completely unsupervised manner. Our algorithm works on motion vectors which can be obtained by partially decoding the compressed video without extracting any additional features. Our approach is based on modelling the motion vector field as a Conditional Random Field (CRF) and obtaining oriented motion segments by finding the optimal labelling which minimises the global energy of CRF. These oriented motion segments are recursively merged based on gradient across their boundaries to obtain the final flow segments. This work in compressed domain can be easily extended to pixel domain by substituting motion vectors with motion based features like optical flow. The proposed algorithm is experimentally evaluated on a standard crowd flow dataset and its superior performance in both accuracy and computational time are demonstrated through quantitative results.

ON THE UTILITY OF CANONICAL CORRELATION ANALYSIS FOR DOMAIN ADAPTATION IN MULTI-VIEW HEADPOSE ESTIMATION

The utility of canonical correlation analysis (CCA) for domain adaptation (DA) in the context of multi-view head pose estimation is examined in this work. We consider the three problems studied in 1], where different DA approaches are explored to transfer head pose-related knowledge from an extensively labeled source dataset to a sparsely labeled target set, whose attributes are vastly different from the source. CCA is found to benefit DA for all the three problems, and the use of a covariance profile-based diagonality score (DS) also improves classification performance with respect to a nearest neighbor (NN) classifier.