A Two-Stage Noise-Tolerant Paradigm for Label Corrupted Person Re-Identification.

Supervised person re-identification (Re-ID) approaches are sensitive to label corrupted data, which is inevitable and generally ignored in the field of person Re-ID. In this paper, we propose a two-stage noise-tolerant paradigm (TSNT) for labeling corrupted person Re-ID. Specifically, at stage one, we present a self-refining strategy to separately train each network in TSNT by concentrating more on pure samples. These pure samples are progressively refurbished via mining the consistency between annotations and predictions. To enhance the tolerance of TSNT to noisy labels, at stage two, we employ a co-training strategy to collaboratively supervise the learning of the two networks. Concretely, a rectified cross-entropy loss is proposed to learn the mutual information from the peer network by assigning large weights to the refurbished reliable samples. Moreover, a noise-robust triplet loss is formulated for further improving the robustness of TSNT by increasing inter-class distances and reducing intra-class distances in the label-corrupted dataset, where a constraint condition for reliability discrimination is carefully designed to select reliable triplets. Extensive experiments demonstrate the superiority of TSNT, for instance, on the Market1501 dataset, our paradigm achieves 90.3% rank-1 accuracy (6.2% improvement over the state-of-the-art method) under noise ratio 20%.

Capra cartilage-derived peptide delivery via carbon nano-dots for cartilage regeneration.

Targeted delivery of site-specific therapeutic agents is an effective strategy for osteoarthritis treatment. The lack of blood vessels in cartilage makes it difficult to deliver therapeutic agents like peptides to the defect area. Therefore, nucleus-targeting zwitterionic carbon nano-dots (CDs) have immense potential as a delivery vehicle for effective peptide delivery to the cytoplasm as well as nucleus. In the present study, nucleus-targeting zwitterionic CDs have been synthesized as delivery vehicle for peptides while also working as nano-agents towards optical monitoring of cartilage healing. The functional groups of zwitterion CDs were introduced by a single-step microwave assisted oxidation procedure followed by COL II peptide conjugation derived from Capra auricular cartilage through NHS/EDC coupling. The peptide-conjugated CDs (PCDs) allows cytoplasmic uptake within a short period of time (∼30 m) followed by translocation to nucleus after ∼24 h. Moreover, multicolor fluorescence of PCDs improves (blue, green, and read channel) its sensitivity as an optical code providing a compelling solution towards enhanced non-invasive tracking system with multifunctional properties. The PCDs-based delivery system developed in this study has exhibited superior ability to induce ex-vivo chondrogenic differentiation of ADMSCs as compared to bare CDs. For assessment of cartilage regeneration potential, pluronic F-127 based PCDs hydrogel was injected to rabbit auricular cartilage defects and potential healing was observed after 60 days. Therefore, the results confirm that PCDs could be an ideal alternate for multimodal therapeutic agents.

Reconstruction Guided Meta-Learning for Few Shot Open Set Recognition.

In many applications, we are constrained to learn classifiers from very limited data (few-shot classification). The task becomes even more challenging if it is also required to identify samples from unknown categories (open-set classification). Learning a good abstraction for a class with very few samples is extremely difficult, especially under open-set settings. As a result, open-set recognition has received limited attention in the few-shot setting. However, it is a critical task in many applications like environmental monitoring, where the number of labeled examples for each class is limited. Existing few-shot open-set recognition (FSOSR) methods rely on thresholding schemes, with some considering uniform probability for open-class samples. However, this approach is often inaccurate, especially for fine-grained categorization, and makes them highly sensitive to the choice of a threshold. To address these concerns, we propose Reconstructing Exemplar-based Few-shot Open-set ClaSsifier (ReFOCS). By using a novel exemplar reconstruction-based meta-learning strategy ReFOCS streamlines FSOSR eliminating the need for a carefully tuned threshold by learning to be self-aware of the openness of a sample. The exemplars, act as class representatives and can be either provided in the training dataset or estimated in the feature domain. By testing on a wide variety of datasets, we show ReFOCS to outperform multiple state-of-the-art methods.

Role of turgor-pressure induced boundary tension in the maintenance of the shoot apical meristem of Arabidopsis thaliana.

In plants, the robust maintenance of tissue structure is crucial to supporting its functionality. The multi-layered shoot apical meristem (SAM) of Arabidopsis, containing stem cells, is an approximately radially symmetric tissue whose shape and structure is maintained throughout the life of the plant. In this paper, a new biologically calibrated pseudo-three-dimensional (P3D) computational model of a longitudinal section of the SAM is developed. It includes anisotropic expansion and division of cells out of the cross-section plane, as well as representation of tension experienced by the SAM epidermis. Results from the experimentally calibrated P3D model provide new insights into maintenance of the structure of the SAM epidermal cell monolayer under tension and quantify dependence of epidermal and subepidermal cell anisotropy on the amount of tension. Moreover, the model simulations revealed that out-of-plane cell growth is important in offsetting cell crowding and regulating mechanical stresses experienced by tunica cells. Predictive model simulations show that tension-determined cell division plane orientation in the apical corpus may be regulating cell and tissue shape distributions needed for maintaining structure of the wild-type SAM. This suggests that cells' responses to local mechanical cues may serve as a mechanism to regulate cell- and tissue-scale patterning.

Evaluating the cell migration potential of TiO2 nanorods incorporated in a Ti6Al4V scaffold: A multiscale approach.

Improvement of cell migration by the nano-topographical modification of implant surface can directly or indirectly accelerate wound healing and osseointegration between bone and implant. Therefore, modification of the implant surface was done with TiO2 nanorod (NR) arrays to develop a more osseointegration-friendly implant in this study. Modulating the migration of a cell, adhered to a scaffold, by the variations of NR diameter, density and tip diameter in vitro is the primary objective of the study. The fluid structure interaction method was used, followed by the submodelling technique in this multiscale analysis. After completing a simulation over a global model, fluid structure interaction data was applied to the sub-scaffold finite element model to predict the mechanical response over cells at the cell-substrate interface. Special focus was given to strain energy density at the cell interface as a response parameter due to its direct correlation with the migration of an adherent cell. The results showed a huge rise in strain energy density after the addition of NRs on the scaffold surface. It also highlighted that variation in NR density plays a more effective role than the variation in NR diameter to control cell migration over a substrate. However, the effect of NR diameter becomes insignificant when the NR tip was considered. The findings of this study could be used to determine the best nanostructure parameters for better osseointegration.

Experimental and simulation investigation of surgical needle insertion into soft tissue mimic biomaterial for minimally invasive surgery (MIS).

The surgical needle insertion process is widely applied in medical interference. During the insertion process, the inhomogeneity and denseness of the soft tissues make it tough to detect the essential tissue damage, a rupture occurs that contains huge forces and material deformations. This study is very important, as all the above-mentioned factors are very significant for modern invasive surgery so that the success rate of the surgery can increase and the patient recovers smoothly. This investigation intends to perform minimally invasive surgical (MIS) procedures and reduce the living tissue damage while performing the biopsy, PCNL, etc. A fracture mechanics method was analyzed to create a needle insertion model which can estimate the needle insertion force during inset in tissue-like PVA gel. The force model was calculated by needle insertion experimentally, and also estimated the needle tip geometry, and diameter influences the fracture toughness. Validate exp. results with simulation results and other papers. It is observed that needle diameter has a significant effect on fracture toughness, whereas the insertion velocity has a slight impact on the fracture toughness. During the rotational needle insertion process, the winds-up of the gel occurs and the diameter of the hole was increasing with increased rpm. Maximum insertion force was noticed in the 27 G needle at 5 mm/s. The interaction function will be less at the maximum fracture development region.

Influence of implant parameters on biomechanical stability of TMJ replacement: A finite element analysis.

The articular disc reduces the stress distribution from the mandible to fossa. In total temporomandibular joint (TMJ) replacement, the implant is required to reduce the stress on fossa implant. Current studies lack standard and optimized parameters for the cylindrical dome on Christensen TMJ implant collar. This study briefed a novel TMJ implant head design and investigates the biomechanical behaviour by considering the articular disc. The radius of the head was varied with the height of the cylinder height to obtain the design of the experiment and the stress distribution was compared with an intact mandible-articular disc model by considering the viscoelastic property of the TMJ disc. The model was simulated at three different angles: 20°, 0° and -20° in the mediolateral direction to simulate the manducation. FEA analysis showed high stresses at the circular heads, and high strength is achieved with increased implant cylinder length and diameter. The results also showed a stress reduction of 50% on the fossa from the mandible. Hence, the newly designed head and suggested modifications may be used as a reference for further clinical improvement of Christensen TMJ as well as other TMJ implants.

Combined computational modeling and experimental analysis integrating chemical and mechanical signals suggests possible mechanism of shoot meristem maintenance.

Stem cell maintenance in multilayered shoot apical meristems (SAMs) of plants requires strict regulation of cell growth and division. Exactly how the complex milieu of chemical and mechanical signals interact in the central region of the SAM to regulate cell division plane orientation is not well understood. In this paper, simulations using a newly developed multiscale computational model are combined with experimental studies to suggest and test three hypothesized mechanisms for the regulation of cell division plane orientation and the direction of anisotropic cell expansion in the corpus. Simulations predict that in the Apical corpus, WUSCHEL and cytokinin regulate the direction of anisotropic cell expansion, and cells divide according to tensile stress on the cell wall. In the Basal corpus, model simulations suggest dual roles for WUSCHEL and cytokinin in regulating both the direction of anisotropic cell expansion and cell division plane orientation. Simulation results are followed by a detailed analysis of changes in cell characteristics upon manipulation of WUSCHEL and cytokinin in experiments that support model predictions. Moreover, simulations predict that this layer-specific mechanism maintains both the experimentally observed shape and structure of the SAM as well as the distribution of WUSCHEL in the tissue. This provides an additional link between the roles of WUSCHEL, cytokinin, and mechanical stress in regulating SAM growth and proper stem cell maintenance in the SAM.

A Micro-Scale Non-Linear Finite Element Model to Optimize the Mechanical Behavior of Bioprinted Constructs.

Extrusion-based bioprinting is an enabling biofabrication technique that is used to create heterogeneous tissue constructs according to patient-specific geometries and compositions. The optimization of bioinks as per requirements for specific tissue applications is an essential exercise in ensuring clinical translation of the bioprinting technologies. Most notably, optimum hydrogel polymer concentrations are required to ensure adequate mechanical properties of bioprinted constructs without causing significant shear stresses on cells. However, experimental iterations are often tedious for optimizing the bioink properties. In this work, a nonlinear finite element modeling approach has been undertaken to determine the effect of different bioink parameters such as composition, concentration on the range of stresses being experienced by the cells in the bioprinted construct. The stress distribution of the cells at different parts of the constructs has also been modeled. It is found that both bioink chemical compositions and concentrations can substantially alter the stress effects experienced by the cells. Concentrated regions of softer cells near pore regions were found to increase stress concentrations by almost three times compared with stress generated in cells away from the pores. The study provides a method for rapid optimization of bioinks, design of bioprinted constructs, as well as toolpath plans for fabricating constructs with homogenous properties.

Text-Based Localization of Moments in a Video Corpus.

Prior works on text-based video moment localization focus on temporally grounding the textual query in an untrimmed video. These works assume that the relevant video is already known and attempt to localize the moment on that relevant video only. Different from such works, we relax this assumption and address the task of localizing moments in a corpus of videos for a given sentence query. This task poses a unique challenge as the system is required to perform: 2) retrieval of the relevant video where only a segment of the video corresponds with the queried sentence, 2) temporal localization of moment in the relevant video based on sentence query. Towards overcoming this challenge, we propose Hierarchical Moment Alignment Network (HMAN) which learns an effective joint embedding space for moments and sentences. In addition to learning subtle differences between intra-video moments, HMAN focuses on distinguishing inter-video global semantic concepts based on sentence queries. Qualitative and quantitative results on three benchmark text-based video moment retrieval datasets - Charades-STA, DiDeMo, and ActivityNet Captions - demonstrate that our method achieves promising performance on the proposed task of temporal localization of moments in a corpus of videos.

Learning Person Re-Identification Models From Videos With Weak Supervision.

Most person re-identification methods, being supervised techniques, suffer from the burden of massive annotation requirement. Unsupervised methods overcome this need for labeled data, but perform poorly compared to the supervised alternatives. In order to cope with this issue, we introduce the problem of learning person re-identification models from videos with weak supervision. The weak nature of the supervision arises from the requirement of video-level labels, i.e. person identities who appear in the video, in contrast to the more precise frame-level annotations. Towards this goal, we propose a multiple instance attention learning framework for person re-identification using such video-level labels. Specifically, we first cast the video person re-identification task into a multiple instance learning setting, in which person images in a video are collected into a bag. The relations between videos with similar labels can be utilized to identify persons, on top of that, we introduce a co-person attention mechanism which mines the similarity correlations between videos with person identities in common. The attention weights are obtained based on all person images instead of person tracklets in a video, making our learned model less affected by noisy annotations. Extensive experiments demonstrate the superiority of the proposed method over the related methods on two weakly labeled person re-identification datasets.

Context-Aware Query Selection for Active Learning in Event Recognition.

Activity recognition is a challenging problem with many practical applications. In addition to the visual features, recent approaches have benefited from the use of context, e.g., inter-relationships among the activities and objects. However, these approaches require data to be labeled, entirely available beforehand, and not designed to be updated continuously, which make them unsuitable for surveillance applications. In contrast, we propose a continuous-learning framework for context-aware activity recognition from unlabeled video, which has two distinct advantages over existing methods. First, it employs a novel active-learning technique that not only exploits the informativeness of the individual activities but also utilizes their contextual information during query selection; this leads to significant reduction in expensive manual annotation effort. Second, the learned models can be adapted online as more data is available. We formulate a conditional random field model that encodes the context and devise an information-theoretic approach that utilizes entropy and mutual information of the nodes to compute the set of most informative queries, which are labeled by a human. These labels are combined with graphical inference techniques for incremental updates. We provide a theoretical formulation of the active learning framework with an analytic solution. Experiments on six challenging datasets demonstrate that our framework achieves superior performance with significantly less manual labeling.

Force modeling to develop a novel method for fabrication of hollow channels inside a gel structure.

Fabrication of hollow channels with user-defined dimensions and patterns inside viscoelastic, gel-type materials is required for several applications, especially in biomedical engineering domain. These include objectives of obtaining vascularized tissues and enclosed or subsurface microfluidic devices. However, presently there is no suitable manufacturing technology that can create such channels and networks in a gel structure. The advent of three-dimensional bioprinting has opened new possibilities for fabricating structures with complex geometries. However, application of this technique to fabricate internal hollow channels in viscoelastic material has not been yet explored to a great extent. In this article, we present the theoretical modeling/background of a proposed manufacturing paradigm through which hollow channels can be conveniently fabricated inside a gel structure. We propose that a tip connected to a robotic arm can be moved in X-, Y-, and Z-axis as per the desired design. The tip can be moved by a magnet or mechanical force. If the tip is further trailed with porous tube and moved inside the viscoelastic material, corresponding internal channels can be fabricated. To achieve this, however, force modeling to understand the forces that will be required to move the tip inside viscoelastic material should be known and understood. Therefore, in our first attempt, we developed the computational force modeling of the tip movement inside gels with different viscoelastic properties to create the channels.

Exploiting Typicality for Selecting Informative and Anomalous Samples in Videos.

In this paper, we present a novel approach to find informative and anomalous samples in videos exploiting the concept of typicality from information theory. In most video analysis tasks, selection of the most informative samples from a huge pool of training data in order to learn a good recognition model is an important problem. Furthermore, it is also useful to reduce the annotation cost as it is time-consuming to annotate unlabeled samples. Typicality is a simple and powerful technique which can be applied to compress the training data to learn a good classification model. In a continuous video clip, an activity shares a strong correlation with its previous activities. We assume that the activity samples that appear in a video form a Markov chain. We explicitly show how typicality can be utilized in this scenario. We compute an atypical score for a sample using typicality and the Markovian property, which can be applied to two challenging vision problems-(a) sample selection for learning activity recognition models, and (b) anomaly detection. In the first case, our approach leads to a significant reduction of manual labeling cost while achieving similar or better recognition performance compared to a model trained with the entire training set. For the latter case, the atypical score has been exploited in identifying anomalous activities in videos where our results demonstrate the effectiveness of the proposed framework over other recent strategies.

Hybrid LSTM and Encoder-Decoder Architecture for Detection of Image Forgeries.

With advanced image journaling tools, one can easily alter the semantic meaning of an image by exploiting certain manipulation techniques such as copy clone, object splicing, and removal, which mislead the viewers. In contrast, the identification of these manipulations becomes a very challenging task as manipulated regions are not visually apparent. This paper proposes a high-confidence manipulation localization architecture that utilizes resampling features, long short-term memory (LSTM) cells, and an encoder-decoder network to segment out manipulated regions from non-manipulated ones. Resampling features are used to capture artifacts, such as JPEG quality loss, upsampling, downsampling, rotation, and shearing. The proposed network exploits larger receptive fields (spatial maps) and frequency-domain correlation to analyze the discriminative characteristics between the manipulated and non-manipulated regions by incorporating the encoder and LSTM network. Finally, the decoder network learns the mapping from low-resolution feature maps to pixel-wise predictions for image tamper localization. With the predicted mask provided by the final layer (softmax) of the proposed architecture, end-to-end training is performed to learn the network parameters through back-propagation using the ground-truth masks. Furthermore, a large image splicing dataset is introduced to guide the training process. The proposed method is capable of localizing image manipulations at the pixel level with high precision, which is demonstrated through rigorous experimentation on three diverse datasets.

Strontium doped hydroxyapatite from Mercenaria clam shells: Synthesis, mechanical and bioactivity study.

Synthesis of strontium-doped hydroxyapatite from Mercenaria clam shells has been carried out by hydrothermal method. The doping of bioceramic, processed from biogenic resources is mostly unexplored. The objective is to understand the effect of strontium (Sr) incorporation on phase stability, sintering behaviour, mechanical properties and cytotoxicity of hydroxyapatite (HAp) derived from clam shells. The different molar concentrations of Sr, varies from 10, 30, 50, 70% of Ca, were substituted into the HAp. The synthesized powders were sintered at 1200 °C in air. The as synthesized powders and sintered specimens were characterized using X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR) and high resolution transmission electron microscopy. The crystallite size and cell parameters of sintered specimens were analyzed from XRD. The XRD of hydrothermally synthesized powders mostly matched with HAp with slight shifting due to Sr doping. However, some distinct Sr based compounds were also observed where Sr substitution is more that 50% of Ca. The XRD of sintered specimen showed increasing ß-tricalcium phosphate (ß-TCP) phase with Sr substitution. The sintered density of solid samples gradually increased from 3.04 g/cc to 3.50 g/cc and surface energy decreased with increasing Sr substitution. Similarly, microhardness, fracture toughness and nanohardness of solid samples found to be enhanced with Sr substitution. The elastic modulus gradually increased from 130 to 137 GPa for HAp and Sr substituted HAp (70% of Ca). The in vitro cytotoxicity of sintered specimen against mouse osteoblast cell line showed that all the samples were nontoxic. However cell proliferation found low for the solid samples containing more than 50% Sr substitution.

Chemometric analysis of integrated FTIR and Raman spectra obtained by non-invasive exfoliative cytology for the screening of oral cancer.

FTIR spectroscopy and Raman spectroscopy of biological analytes are increasingly explored as screening tools for early detection of cancer. In the present study, an integrated analysis of the FTIR and Raman spectra obtained from exfoliated cells is adopted to improve discrimination of normal, pre-cancerous and cancerous conditions. Multiple spectra were obtained from 13 normal, 13 pre-cancer and 10 cancer patients in both modes. Compared to normal patients, significant differences were observed at 1550, 1580, 1640, 2370, 2330, 2950-3000 and 3650-3750 cm-1 (FTIR) and 520, 640, 785, 827, 850, 935, 1003, 1175, 1311 cm-1 and 1606 cm-1 (Raman) vibrations of the other two. The increase in DNA, protein and lipid content with malignancy was more clearly elucidated by examining both spectra. Principal component analysis (PCA)-linear discriminant analysis (LDA) with 10-fold cross validation of the FTIR and Raman spectral data sets showed efficient discrimination between normal and pathological conditions while overlapping was seen between the two pathologies. The PCA-LDA model of the dual spectra yielded a classification accuracy of 98% in comparison with either FTIR (85%) or Raman (82%) in a spectrum-wise comparison. In the patient-wise approach (mean of all spectra from a patient), the overall classification efficiency was 73%, 80% and 87% for FTIR, Raman and integrated spectral approaches respectively. Moreover, the efficiency of the integrated FTIR-Raman PCA-LDA model as a prediction tool was tested to screen susceptible individuals (11 cigarette smokers) using the dual spectra acquired from these individuals. The study presents proof-of-concept for adopting a large-scale, follow-up trial of the approach for mass screening purposes.

Diversity-Aware Multi-Video Summarization.

Most video summarization approaches have focused on extracting a summary from a single video; we propose an unsupervised framework for summarizing a collection of videos. We observe that each video in the collection may contain some information that other videos do not have, and thus exploring the underlying complementarity could be beneficial in creating a diverse informative summary. We develop a novel diversity-aware sparse optimization method for multi-video summarization by exploring the complementarity within the videos. Our approach extracts a multi-video summary, which is both interesting and representative in describing the whole video collection. To efficiently solve our optimization problem, we develop an alternating minimization algorithm that minimizes the overall objective function with respect to one video at a time while fixing the other videos. Moreover, we introduce a new benchmark data set, Tour20, that contains 140 videos with multiple manually created summaries, which were acquired in a controlled experiment. Finally, by extensive experiments on the new Tour20 data set and several other multi-view data sets, we show that the proposed approach clearly outperforms the state-of-the-art methods on the two problems-topic-oriented video summarization and multi-view video summarization in a camera network.

Optimal Landmark Selection for Registration of 4D Confocal Image Stacks in Arabidopsis.

Technologically advanced imaging techniques have allowed us to generate and study the internal part of a tissue over time by capturing serial optical images that contain spatio-temporal slices of hundreds of tightly packed cells. Image registration of such live-imaging datasets of developing multicelluar tissues is one of the essential components of all image analysis pipelines. In this paper, we present a fully automated 4D(X-Y-Z-T) registration method of live imaging stacks that takes care of both temporal and spatial misalignments. We present a novel landmark selection methodology where the shape features of individual cells are not of high quality and highly distinguishable. The proposed registration method finds the best image slice correspondence from consecutive image stacks to account for vertical growth in the tissue and the discrepancy in the choice of the starting focal point. Then, it uses local graph-based approach to automatically find corresponding landmark pairs, and finally the registration parameters are used to register the entire image stack. The proposed registration algorithm combined with an existing tracking method is tested on multiple image stacks of tightly packed cells of Arabidopsis shoot apical meristem and the results show that it significantly improves the accuracy of cell lineages and division statistics.

Context-Aware Surveillance Video Summarization.

We present a method that is able to find the most informative video portions, leading to a summarization of video sequences. In contrast to the existing works, our method is able to capture the important video portions through information about individual local motion regions, as well as the interactions between these motion regions. Specifically, our proposed Context-Aware Video Summarization (CAVS) framework adopts the methodology of sparse coding with generalized sparse group lasso to learn a dictionary of video features and a dictionary of spatio-temporal feature correlation graphs. Sparsity ensures that the most informative features and relationships are retained. The feature correlations, represented by a dictionary of graphs, indicate how motion regions correlate to each other globally. When a new video segment is processed by CAVS, both dictionaries are updated in an online fashion. Specifically, CAVS scans through every video segment to determine if the new features along with the feature correlations, can be sparsely represented by the learned dictionaries. If not, the dictionaries are updated, and the corresponding video segments are incorporated into the summarized video. The results on four public datasets, mostly composed of surveillance videos and a small amount of other online videos, show the effectiveness of our proposed method.