JSCC-Cast: A Joint Source Channel Coding Video Encoding and Transmission System with Limited Digital Metadata

This work considers the design and practical implementation of JSCC-Cast, a comprehensive analog video encoding and transmission system requiring a reduced amount of digital metadata. Suitable applications for JSCC-Cast are multicast transmissions over time-varying channels and Internet of Things wireless connectivity of end devices having severe constraints on their computational capabilities. The proposed system exhibits a similar image quality compared to existing analog and hybrid encoding alternatives such as Softcast. Its design is based on the use of linear transforms that exploit the spatial and temporal redundancy and the analog encoding of the transformed coefficients with different protection levels depending on their relevance. JSCC-Cast is compared to Softcast, which is considered the benchmark for analog and hybrid video coding, and with an all-digital H.265-based encoder. The results show that, depending on the scenario and considering image quality metrics such as the structural similarity index measure, the peak signal-to-noise ratio, and the perceived quality of the video, JSCC-Cast exhibits a performance close to that of Softcast but with less metadata and not requiring a feedback channel in order to track channel variations. Moreover, in some circumstances, the JSCC-Cast obtains a perceived quality for the frames comparable to those displayed by the digital one.

Yeast cell fate control by temporal redundancy modulation of transcription factor paralogs

Recent single-cell studies have revealed that yeast stress response involves transcription factors that are activated in pulses. However, it remains unclear whether and how these dynamic transcription factors temporally interact to regulate stress survival. Here we show that budding yeast cells can exploit the temporal relationship between paralogous general stress regulators, Msn2 and Msn4, during stress response. We find that individual pulses of Msn2 and Msn4 are largely redundant, and cells can enhance the expression of their shared targets by increasing their temporal divergence. Thus, functional redundancy between these two paralogs is modulated in a dynamic manner to confer fitness advantages for yeast cells, which might feed back to promote the preservation of their redundancy. This evolutionary implication is supported by evidence from Msn2/Msn4 orthologs and analyses of other transcription factor paralogs. Together, we show a cell fate control mechanism through temporal redundancy modulation in yeast, which may represent an evolutionarily important strategy for maintaining functional redundancy between gene duplicates.

Elastic AlignedSENSE for Dynamic MR Reconstruction: A Proof of Concept in Cardiac Cine

Numerous methods in the extensive literature on magnetic resonance imaging (MRI) reconstruction exploit temporal redundancy to accelerate cardiac cine. Some of them include motion compensation, which involves high computational costs and long runtimes. In this work, we proposed a method—elastic alignedSENSE (EAS)—for the direct reconstruction of a motion-free image plus a set of nonrigid deformations to reconstruct a 2D cardiac sequence. The feasibility of the proposed approach was tested in 2D Cartesian and golden radial multi-coil breath-hold cardiac cine acquisitions. The proposed approach was compared against parallel imaging compressed sense (sPICS) and group-wise motion corrected compressed sense (GWCS) reconstructions. EAS provides better results on objective measures with considerable less runtime when an acceleration factor is higher than 10×. Subjective assessment of an expert, however, invited proposing the combination of EAS and GWCS as a preferable alternative to GWCS or EAS in isolation.

Attention-Based Temporal Encoding Network with Background-Independent Motion Mask for Action Recognition

Convolutional neural network (CNN) has been leaping forward in recent years. However, the high dimensionality, rich human dynamic characteristics, and various kinds of background interference increase difficulty for traditional CNNs in capturing complicated motion data in videos. A novel framework named the attention-based temporal encoding network (ATEN) with background-independent motion mask (BIMM) is proposed to achieve video action recognition here. Initially, we introduce one motion segmenting approach on the basis of boundary prior by associating with the minimal geodesic distance inside a weighted graph that is not directed. Then, we propose one dynamic contrast segmenting strategic procedure for segmenting the object that moves within complicated environments. Subsequently, we build the BIMM for enhancing the object that moves based on the suppression of the not relevant background inside the respective frame. Furthermore, we design one long-range attention system inside ATEN, capable of effectively remedying the dependency of sophisticated actions that are not periodic in a long term based on the more automatic focus on the semantical vital frames other than the equal process for overall sampled frames. For this reason, the attention mechanism is capable of suppressing the temporal redundancy and highlighting the discriminative frames. Lastly, the framework is assessed by using HMDB51 and UCF101 datasets. As revealed from the experimentally achieved results, our ATEN with BIMM gains 94.5% and 70.6% accuracy, respectively, which outperforms a number of existing methods on both datasets.

Robust Computing for Machine Learning-Based Systems

The drive for automation and constant monitoring has led to rapid development in the field of Machine Learning (ML). The high accuracy offered by the state-of-the-art ML algorithms like Deep Neural Networks (DNNs) has paved the way for these algorithms to being used even in the emerging safety-critical applications, e.g., autonomous driving and smart healthcare. However, these applications require assurance about the functionality of the underlying systems/algorithms. Therefore, the robustness of these ML algorithms to different reliability and security threats has to be thoroughly studied and mechanisms/methodologies have to be designed which result in increased inherent resilience of these ML algorithms. Since traditional reliability measures like spatial and temporal redundancy are costly, they may not be feasible for DNN-based ML systems which are already super computer and memory intensive. Hence, new robustness methods for ML systems are required. Towards this, in this chapter, we present our analyses illustrating the impact of different reliability and security vulnerabilities on the accuracy of DNNs. We also discuss techniques that can be employed to design ML algorithms such that they are inherently resilient to reliability and security threats. Towards the end, the chapter provides open research challenges and further research opportunities.

Redundancy Circuits of the Commissural Pathways in Human and Rhesus Macaque Brains

It has been hypothesized that the human brain has traded redundancy for efficiency, but the structural existence has not been identified to examine this claim. Here, we report three redundancy circuits of the commissural pathways in primate brains, namely the orbitofrontal, temporal, and occipital redundancy circuits of the anterior commissure and corpus callosum. Each redundancy circuit has two distinctly separated routes connecting a common pair of cortical regions. We mapped their trajectories in human and rhesus macaque brains using individual and population-averaged tractography. The dissection results confirmed the existence of these redundancy circuits connecting the orbitofrontal lobe, amygdala, and visual cortex. The volume analysis showed a significant reduction in the orbitofrontal and occipital redundancy circuits of the human brain, whereas the temporal redundancy circuit had a substantial organizational difference between the human and rhesus macaque. Our overall findings suggest that the human brain is more efficient in the commissural pathway, as shown by the significantly reduced volume of the anterior commissure which serves as the backup connections for the corpus callosum. This reduction of the redundancy circuit may explain why humans are more vulnerable to psychiatric brain disorders stemming from the corpus callosum compared to non-human primates.SignificanceWe report and describe the connection routes of three redundancy circuits of the commissural pathways in human and rhesus macaque brains and compare their volumes. Our tractography and dissection studies confirmed that the human brain has smaller redundancy circuits. This is the first time such redundancy circuits of the commissural pathways have been identified, and their differences quantified in human and rhesus macaque to verify the redundancy-efficiency tradeoff hypothesis. The findings provide new insight into the topological organization of the human brain and may help understand the circuit mechanism of brain disorders involving these pathways.

Video Compression for Surveillance Application using Deep Neural Network

We present a new approach to video compression for video surveillance by refining the shortcomings of conventional approach and substitute each traditional component with their neural network counterpart. Our proposed work consists of motion estimation, compression and compensation and residue compression, learned end-to-end to minimize the rate-distortion trade off. The whole model is jointly optimized using a single loss function. Our work is based on a standard method to exploit the spatio-temporal redundancy in video frames to reduce the bit rate along with the minimization of distortions in decoded frames. We implement a neural network version of conventional video compression approach and encode the redundant frames with lower number of bits. Although, our approach is more concerned toward surveillance, it can be extended easily to general purpose videos too. Experiments show that our technique is efficient and outperforms standard MPEG encoding at comparable bitrates while preserving the visual quality.