Abstract

The popularity of egocentric cameras and their always-on nature has lead to the abundance of day long first-person videos. The highly redundant nature of these videos and extreme camera-shakes make them difficult to watch from beginning to end. These videos require efficient summarization tools for consumption. However, traditional summarization techniques developed for static surveillance videos or highly curated sports videos and movies are either not suitable or simply do not scale for such hours long videos in the wild. On the other hand, specialized summarization techniques developed for egocentric videos limit their focus to important objects and people. This paper presents a novel unsupervised reinforcement learning framework to summarize egocentric videos both in terms of length and the content. The proposed framework facilitates incorporating various prior preferences such as faces, places, or …

Abstract

We consider a generalization of the recently proposed gradient coding framework where a large dataset is divided across n workers and each worker transmits to a master node one or more linear combinations of the gradients over the data subsets assigned to it. Unlike the conventional framework which requires the master node to recover the sum of the gradients over all the data subsets in the presence of s straggler workers, we relax the goal of the master node to computing the sum of at least some α fraction of the gradients. The broad goal of our work is to study the optimal computation and communication load per worker for this approximate gradient coding framework. We begin by deriving a lower bound on the computation load of any feasible scheme and also propose a strategy which achieves this lower bound, albeit at the cost of high communication load and a number of data partitions which can be polynomial in the number of workers n. We then restrict attention to schemes which utilize a number of data partitions equal to n and propose schemes based on cyclic assignment which have a lower communication load. When each worker transmits a single linear combination, we also prove lower bounds on the computation load of any scheme using n data partitions.

Abstract

A smart thermal patch for adaptive thermotherapy is provided. In an embodiment, the patch can be a stretchable, non-polymeric, conductive thin film flexible and non-invasive body integrated mobile thermal heater with wireless control capabilities that can be used to provide adaptive thermotherapy. The patch can be geometrically and spatially tunable on various pain locations. Adaptability allows the amount of heating to be tuned based on the temperature of the treated portion.

Abstract

The emergence of vehicle connectivity technologies and associated applications have paved the way for increased consumer interest in connected vehicles. These modern day vehicles are now capable of sending/receiving vast amounts of data and offloading computation (which is one possible service) to servers thereby improving safety, comfort, driving experience, etc. In the early stages of connectivity, all the data communication and computation offloading happened between the cloud server and the vehicles. However, this is not feasible in scenarios having strict timing requirements and bandwidth cost constraints. Vehicular Edge Computing (VEC) demonstrated an efficient way to tackle the above problem. In order to optimally utilize the resources of the edge servers for data delivery, an efficient edge resource allocation framework needs to be developed. In a recent work, data/service delivery to connected vehicles assumed a worst-case scenario that all vehicles with routes passing through an edge appear in the edge coverage region simultaneously. However, this worst-case scenario is very pessimistic, which results in overestimation of edge resources. We address this by precisely computing the set of vehicles which simultaneously appear in the coverage region of an edge (which we call vehicle overlaps). In this work, we first propose an optimization framework for edge resource allocation that minimizes the bandwidth cost of data delivery to connected vehicles while considering the traffic flow and vehicle overlaps. Then, we propose an efficient heuristic to deliver data based on minimizing global edge bandwidth cost gradient under vehicle overlaps. We demonstrate the improvement in resource allocation considering vehicle overlaps. Using real world traffic data, we also demonstrate reduction in data delivery times using the proposed heuristic.

Abstract

A processor-implemented method for generating a lip-sync for a face to a target speech of a live session to a speech in one or more languages in-sync with improved visual quality using a machine learning model and a pre-trained lip-sync model is provided. The method includes (i) determining a visual representation of the face and an audio representation, the visual representation includes crops of the face; (ii) modifying the crops of the face to obtain masked crops; (iii) obtaining a reference frame from the visual representation at a second timestamp; (iv) combining the masked crops at the first timestamp with the reference to obtain lower half crops; (v) training the machine learning model by providing historical lower half crops and historical audio representations as training data; (vi) generating lip-synced frames for the face to the target speech, and (vii) generating an in-sync lip-synced frames by the pre-trained lip-sync model.

Abstract

Standard Model Predictive Control (MPC) or trajectory optimization approaches perform only a local search to solve a complex non-convex optimization problem. As a result, they cannot capture the multi-modal characteristic of human driving. A global optimizer can be a potential solution but is computationally intractable in a real-time setting. In this paper, we present a real-time MPC capable of searching over different driving modalities. Our basic idea is simple: we run several goaldirected parallel trajectory optimizations and score the resulting trajectories based on user-defined meta cost functions. This allows us to perform a search over several locally optimal motion plans. Although conceptually straightforward, realizing this idea in real-time with existing optimizers is highly challenging from technical and computational standpoints. With this motivation, we present a novel batch non-holonomic trajectory optimization whose underlying matrix algebra is easily parallelizable across problem instances and reduces to computing large batch matrix-vector products. This structure, in turn, is achieved by deriving a linearization-free multi-convex reformulation of the non-holonomic kinematics and collision avoidance constraints. We extensively validate our approach using both synthetic and real data sets (NGSIM) of traffic scenarios. We highlight how our algorithm automatically takes lane-change and overtaking decisions based on the defined meta cost function. Our batch optimizer achieves trajectories with lower meta cost, up to 6x faster than competing baselines.

Abstract

Transfer learning is a popular method for obtaining deep trained models for data-scarce face tasks such as head pose and emotion. However, current transfer learning methods are inefficient and time-consuming as they do not fully account for the relationships between related tasks. Moreover, the transferred model is large and computationally expensive. As an alternative, we propose ETL: a technique that efficiently transfers a pre-trained model to a new task by retaining only cross-task aware filters, resulting in a sparse transferred model. We demonstrate the effectiveness of ETL by transferring VGGFace, a popular face recognition model to four diverse face tasks. Our experiments show that we attain a size reduction up to 97% and an inference time reduction up to 94% while retaining 99.5% of the baseline transfer learning accuracy

Abstract

Quantum sensing applications such as nitrogen vacancy based magnetometry require phase locked loops (PLL), which can synthesize microwave frequencies in a narrow band near 2.87 GHz. Moreover, it is also required that the PLLs should have low noise and low jitter for high stability and fast settling time. These requirements seek low phase noise voltage controlled oscillator (VCO) with small variation in its gain (KVCO) within the desired tuning range. In this paper, we present a technique for designing a low phase noise VCO with low KVCO and small KVCO variation. To validate the proposed technique, an LC VCO has been designed and implemented in 0.18 µm CMOS process and post layout simulation results are presented. The simulation results show that the proposed LC VCO achieves KVCO variation of ±3.82% in the frequency range of 2.75 - 2.94 GHz and exhibits a phase noise of -118.64 dBc/Hz at an offset of 1 MHz, while consuming 9 mW of power from a 1.8 V supply.

Abstract

Precision farming has a positive potential in the agricultural industry regarding water conservation, increased productivity, better development of rural areas, and increased income. Blockchain technology is a better alternative for storing and sharing farm data as it is reliable, transparent, immutable, and decentralized. Remote monitoring of an agricultural field requires security systems to ensure that any sensitive information is exchanged only among authenticated entities in the network. To this end, we design an efficient blockchain-enabled authenticated key agreement scheme for mobile vehicles-assisted precision agricultural Internet of Things (IoT) networks called AgroMobiBlock . The limited existing work on authentication in agricultural networks shows passive usage of blockchains with very high costs. AgroMobiBlock proposes a novel idea using the elliptic curve operations on an active hybrid blockchain over mobile farming vehicles with low computation and communication costs. Formal and informal security analysis along with the formal security verification using the Automated Validation of Internet Security Protocols and Applications (AVISPA) software tool have shown the robustness of AgroMobiBlock against man-in-the-middle, impersonation, replay, physical capture, and ephemeral secret leakage attacks among other potential attacks. The blockchain-based simulation on large-scale nodes shows the computational time for an increase in the network and block sizes. Moreover, the real-time testbed experiments have been performed to show the practical usefulness of the proposed scheme.

Abstract

Coral Reefs are vital ecosystems as they support a vast population of aquatic life and help in maintaining biodiversity. Currently, these ecosystems are under severe threat due to rising ocean temperatures and contaminated surroundings. Hence, Coral Reefs need to be monitored regularly to study their health and protect them in a better way. Remote detection and assessment of Coral Reefs using satellite data would significantly enhance the reach and ensure continuous Coral Reef monitoring. The spectral signature of the Coral Reefs captured by the satellite sensor could be a great asset in this process. Still, these signatures are heavily dopped with several unwanted noise signals coming from various surface and atmospheric features. It is essential to remove these noise signals from the satellite data. This paper studied the effectiveness of different available atmospheric correction methods such as QUAC, 6S, and FLAASH Atmospheric correction tools (ACT) in removing atmospheric noise and studied the impact of sun glint and water column correction on improving remote detection of Coral Reefs. As a case study, these processes were applied on Landsat ETM+ to study the abundant Coral habitat near the North Reef Island at the Andaman Islands. The results indicated that FLAASH is the best atmospheric correction tool among the selected algorithms as FLAASH corrected image facilitated better identification of Coral Reefs. Hence this image was further enhanced using the Hedley sun glint and Lyzenga water column correction methods which improved Coral Reef detectability significantly. The detection of Corals from the image was done using ISO-DATA unsupervised classification to evaluate the role of each process in enhancing the Coral detection. Visual inspection was used to assess the classifier accuracy towards Coral reef detection. The results showed that a combination of FLAASH, sun glint removal, and water column correction significantly improved Coral Reef identification from the remote sensing data.