Search

Fully immersive experiences in AR/VR depend on re-constructing the full body pose of the user without restricting their motion. In this paper we study the use of body-worn electromagnetic (EM) field-based sensing for the task of 3D human pose reconstruction. To this end, we present a method to estimate SMPL parameters from 6-12 EM sensors. We leverage a customized wearable system consisting of wireless EM sensors measuring time-synchronized 6D poses at 120 Hz. To provide accurate poses even with little user instrumentation, we adopt a recently proposed hybrid framework, learned gradient descent (LGD), to iteratively estimate SMPL pose and shape from our input measurements. This allows us to harness powerful pose priors to cope with the idiosyncrasies of the input data and achieve accurate pose estimates. The proposed method uses AMASS to synthesize virtual EM-sensor data and we show that it generalizes well to a newly captured real dataset consisting of a total of 36 minutes of motion from 5 subjects. We achieve reconstruction errors as low as 31.8 mm and 13.3 degrees, outperforming both pure learning- and pure optimization-based methods. Code and data is available under https://ait.ethz.ch/projects/2021/em-pose., 2021 IEEE/CVF International Conference on Computer Vision (ICCV), ISBN:978-1-6654-2812-5

The Sargasso Sea has been a ocean life habitat for millions of years, yet accurate assessment of the boundary area and detection of these relatively small sea surface features using Landsat series and Moderate resolution Imaging Spectroradiometer (MODIS) instruments have been found to have difficulty or even impossible due to lack of spatial resolution, coverage, recurring observance, and algorithm limitations to identify pelagic species of Sargassum. Sargassum rafts tend to be elongated, curved in the upwind direction and warmer than the surrounding ocean surface. Long weed ‘trails’ extending upwind from the rafts are evidence of plants dropping out and being left behind. Satellite data utilizing a simple ocean color indexes such as the floating algae Index and Normalized Differential Vegetation Index (NDVI) have been established to detect floating algae in open environments using MODIS instruments. Floating Algae Index (FAI) has shown advantages over the traditional NDVI and Enhanced Vegetation Index (EVI) because FAI is less sensitive to changes in environmental and observing conditions (aerosol type and thickness, solar/viewing geometry, and sun glint) and can see through thin clouds. The baseline subtraction method provides a simple yet effective means for atmospheric correction. The algorithms assisted in identifying the boundary area of the Sargasso Sea and the path of this floating alga past Cape Hatteras into the Atlantic Ocean. Due to the fact that similar spectral bands are available on many existing and planned satellite sensors such as Landsat series observations satellites, the NDVI and FIA concept was extendable to establish a long-term record of these ecologically biological dependent ocean plants.

Skin stretch is a novel haptic feedback method that can provide a human operator with information about the magnitude and direction of an applied force. To evaluate the potential for skin stretch feedback to be used as a sensory substitute for kinesthetic (force) feedback in robotic teleoperation systems, a study was conducted to measure the ability of users to discriminate environment stiffness using varying levels of fingerpad skin stretch instead of force feedback. A new, high-fidelity skin stretch feedback device was developed that imposes tangential fingerpad skin stretch in proportion to the intended level of force feedback. In psychophysical experiments, users received skin stretch feedback with magnitude proportional to the users' penetration depth into a virtual wall. Users' stiffness discrimination capability using skin stretch was comparable to that of using force feedback. Furthermore, larger skin stretch cues were perceived by users as portraying greater stiffness without any advance training, which indicates that skin stretch feedback would be an intuitive sensory substitute for force feedback. Thus, skin stretch feedback is a promising method for conveying kinesthetic force information in applications such as robot-assisted surgery, where high levels of force feedback may not be desirable due to stability or safety concerns.

Autonomous Underwater Vehicles (AUVs) often communicate with scientists on the surface over an unreliable acoustic channel. The challenges of operating in deep waters, over long distances, and with surface ship noise amount to a communication channel with a very low effective bandwidth. This restriction makes transmission of images, even highly compressed images, quite difficult. We present an image compression algorithm designed to convey the gist of an image to surface operators in a very small number of bytes. Our technique divides a large existing database of underwater images into ‘tiles’, and uses these to reconstruct an approximation to new underwater images from a similar domain. We achieve significantly higher compression ratios than conventional image compression techniques, such as JPEG or SPIHT, while still being able to provide useful visual feedback to the surface.

In this work, we propose a random access protocol for decentralized detection in wireless sensor networks. In this scheme, sensors are grouped according to the informativeness of their data. Then fusion center collects sensor data sequentially in the order of their informativeness and terminates the fusion process once the target performance is reached. To determine the optimal transmission probability in random access, we propose a novel Bayesian update algorithm utilizing both the sensing information and the channel feedback. As shown by our simulations, incorporating sensing information greatly improves the communication efficiency over a generic Bayesian update scheme relying only on channel feedback. Comparison with fixed sample size test and sequential probability ratio test shows that the proposed scheme achieves significant channel and power efficiency gain over existing strategies. Index Terms—Sensor networks, Decentralized Detection, Mul- tiaccess Communication, Sequential Detection.

A 5-GHz CMOS fractional-N frequency synthesizer with a delta-sigma modulator is designed in this paper. The frequency dividers are composed of an injection-locked frequency divider and a programmable divider. In consideration of low power consumption, we use an injection-locked frequency divider as the first stage prescaler. The loop filter is a second-order passive filter. The delta-sigma modulator is MASH 1-1-1 architecture. The VCO exhibits a phase noise of-116dBc/Hz at 1MHz offset frequency and an output frequency ranges from 4.91GHz to 5.38GHz. TSMC 0.18-mum CMOS process is used for this frequency synthesizer design and simulation. The frequency resolution of this fractional-N frequency synthesizer is 27 KHz and the locking time is 8 mus.

Bioelectronic interfaces that allow dehydrogenase enzymes to electrically communicate with electrodes have potential applications in the development of biosensors and biocatalytic reactors. A fully scalable, post-CMOS-compatible, three-electrode interface to biochemical sensors, consisting of Ti/Au working and auxiliary electrodes and a Ti/Au/Ag/AgCl reference electrode, has been developed. Also described is a tri-functional linking molecule that binds the mediator and cofactor to the electrode in a unique spatial arrangement in which the dehydrogenase enzyme can bind to cofactor and multistep electron transfer between the electrode and enzyme is achieved. This approach provides greater flexibility in assembling complex bioelectronic interfaces than is possible with previously reported, linear linking molecules. A cysteine molecule was self-assembled on a gold electrode via a thiol bond. The electron mediator toluidine blue O (TBO) and the cofactor, /spl beta/-nicotinamide adenine dinucleotide phosphate (NADP/sup +/) were chemically attached to cysteine via the formation of amide bonds. Cyclic voltammetry, was used to demonstrate the electrical activity, and enzymatic activity of the resulting bioelectronic interface.

A multi-analyte electrochemical sensor array platform has been developed for protein-based biosensors utilizing post-CMOS compatible fabrication procedures that enables formation of single-chip biosensor array microsystems. The process was developed on a glass substrate to emulate the surface of a CMOS chip. A three-electrode system, including an array of gold working electrodes (WE) and an on-chip Ag/AgCl reference electrode (RE), was formed to facilitate electrochemical analysis protein-based interfaces. SU-8 epoxy resin was subsequently applied and patterned as an insulation and planerization layer before application of microfluidic channels used to self-assemble novel bio-interfaces on individual working electrodes. The protein-based bio-interfaces provide selectivity, and the electrode floorplan maximizes sensitivity and minimizes solution resistance errors. The electrodes on the prototype platform can be easily scaled to form ~100 electrodes on the surface of a 10 mm2 CMOS chip.

In this report, the transparent conducting indium zinc oxide (IZO) films (60-220 nm) have been grown on hardness poly-carbonate (HPC) substrate without a post deposition annealing treatment. The direct current (dc) magnetron sputtering system was employed for the film deposition. The IZO alloy target (99.99% and 95% in purity and density, respectively) is composed of 90 wt.% of ln2O3 and 10 wt.% of ZnO. The electrical, optical, and structural properties of these prepared films by different dc powers, such as 50 W, 80 W and 100 W, without/with the ion-assisted deposition (IAD) technique. An optimum IZO deposition condition is developed for flexible organic light-emitting device (OLED) applications. The IZO films grown at low temperature (~ 50degC) by the dc magnetron sputtering (100-W power) with the IAD technique were used to study the electroluminescence (EL) performance of OLEDs. Under a current density of 200 (mA/cm2), the developed OLED/IZO/HPC substrate shows an excellent efficiency (5 V turn-on voltage) and a luminance of 1800 (cd/m2) in average, which is better than that measured with commercial indium-tin oxide (ITO) anodes and well above the electro-optical application standard.

The hydrogenated poly-silicon germanium (poly-SiGe:H) epitaxial film has been investigated by gold induced lateral crystallization (Au-ILC) technology on a-SiGe:H layer at 10-hr 350degC annealing temperature and 60-sccm hydrogen (H2) content. By this optimal condition, the growth rate by Au induced can be as large as 15.9 mum/hr. Due to low annealing temperature treatment (les 400degC) and large growth rate, this novel technology will be a noticeable poly-SiGe:H pin IR-sensing fabrication on a conventional pre-coated indium tin oxide (ITO)-glass substrate. Under 1-muW IR-LED incident (with peak wavelength at 710 nm) and at 5-V biased voltage, the poly-SiGe:H pin IR sensor developed by the Au-ILC technology, i.e., an Al (anode)/n poly-SiGe:H/i poly-SiGe:H/p poly-SiGe:H/ITO (cathode)/glass-substrate structure posses a maximum optical gain and response speed, almost 600% and 130%, respectively, better than that of a traditional pin type. Meanwhile, the FWHM of a poly-SiGe:H pin sensor with Au-ILC technology corresponding to a traditional pin sensor can be reduced from 280 to 150 nm, thus ascertaining its good IR-sensing selectivity. These better IR-sensing performances are demonstrated again that the proposed Au-ILC technology is a candidate to the low cost IC on opto-electronic applications.

We describe a new VNIR-SWIR atmospheric correction method for multi- and hyperspectral imagery, dubbed QUAC (QUick Atmospheric Correction) that also enables retrieval of the wavelength-dependent optical depth of the aerosol or haze and molecular absorbers. It determines the atmospheric compensation parameters directly from the information contained within the scene using the observed pixel spectra. The approach is based on the empirical finding that the spectral standard deviation of a collection of diverse material spectra, such as the endmember spectra in a scene, is essentially spectrally flat. It allows the retrieval of reasonably accurate reflectance spectra even when the sensor does not have a proper radiometric or wavelength calibration, or when the solar illumination intensity is unknown. The computational speed of the atmospheric correction method is significantly faster than for the first-principles methods, making it potentially suitable for realtime applications. The aerosol optical depth retrieval method, unlike most prior methods, does not require the presence of dark pixels. In this paper, QUAC is applied to atmospherically correction several AVIRIS data sets. Comparisons to the physics-based FLAASH code are also presented.

This paper presents an integrated, protein-based, biosensor that can be scaled to form high-density, multi-analyte sensor arrays physically integrated on a signal conditioning circuit die. A fully scalable, post-CMOS-compatible, three-electrode interface to biochemical sensors has been developed. A silicon substrate electrode system, consisting of Ti/Au working and auxiliary electrodes and a Ti/Au/Ag/AgCl reference electrode has been adapted to biomimetic sensors. The functional Ag/AgCl reference electrode is isolated from the environment using a Nafion cation-exchange membrane to extend operation lifetime. To complete the sensor structure, lipid bilayers have been deposited in passivation layer openings formed over individual working electrodes using a special tethering molecule. Total internal reflection microscopy (TIRFM) studies were done to confirm that a wide range of proteins, such as dehydrogenase enzymes and ion channels, can then be embedded into the lipid bilayers. These results verify the potential to form highly selective recognition elements with direct physical connection to readout electronics on the supporting silicon substrate.

Our previous work is presented on diversity antenna experiments in the 2.3 GHz band, specifically for applications in the automotive environment. We describe the case of two-element position diversity, and we study the effects of the antenna positions on the amount of diversity gain. The presentation consists primarily of measurement data on a moving vehicle, with software post-processing to determine the effects of diversity combining. We present several diversity combining schemes, and describe a switched diversity circuit that we have built to address one of these schemes. An in-house developed SPDT MEMS switch, which has desirable insertion loss and isolation characteristics for the diversity switch, is shown. Also, a discussion on using a switch for the switched diversity antenna is included.

We describe RF MEMS switches and their applications to tunable filters and switchable antenna elements. We describe different scenarios where RF photonic systems can benefit from RF MEMS device development.

Although modified lead-titanate ceramics exhibit a negligible d/sub 31/ piezoelectric coefficient, piezoelectric-rod/polymer-matrix composites made from them exhibit a substantial d/sub 31/. A theoretical analysis shows that the composite's enhanced d/sub 31/ coefficient arises from lateral stress on the polymer phase being transferred to a longitudinal stress along the ceramic rods by the Poisson effect in the polymer, thus producing a charge through the ceramic's d/sub 33/. For hydrophone applications, this enhanced d/sub 31/ suppresses the hydrostatic response so that the composite's hydrophone figure of merit, d/sub h/g/sub h/, is less than that of the constituent ceramic. Nevertheless, the composite's still substantial d/sub h/g/sub h/, high g/sub h/ coefficient, remarkable pressure stability, low density, formability, and availability in thick sheets provide advantages for naval applications. In ultrasonic transducer applications, the composite structure provides no enhancement in electromechanical coupling, since the modified lead-titanate's k/sub 33/ is nearly the same as its k/sub t/. These expectations are in good agreement with measurements on 1-3 modified lead-titanate/polymer composites made by the dice-and-fill technique. >

To provide improved input/output performance, as well as to accommodate growing storage requirements, disk subsystems are increasing in size. A major hurdle to obtaining the performance available from these large disk subsystems is load imbalance, or disk skew. Dynamic data placement, the conventional load balancing technique, is usually adequate to deal with load imbalance because it is forced to accept atomic data sets with rapidly changing access patterns. The authors name this rapid fluctuation 'floating load imbalance' and distinguish it from the conventional view of load imbalance, referred to as 'fixed load imbalance'. Disk striping at a high granularity is suggested as a solution to floating load imbalance, the atomic data set problem and the complexity of balancing large disk subsystems. Disk striping uniformly spreads data sets across the disks in the subsystem and essentially randomizes, the disk accessed by each request. This randomization effectively handles both fixed and floating load imbalance. >

When a disk in the RAID5 disk array architecture has failed, requests to that disk can only be serviced by reading data from all surviving disks and rebuilding the lost data. This may cut disk performance in half. To avoid this degradation, all of the lost data must be rebuilt and written to a spare disk. The faster the data are rebuilt, the sooner the disk array returns to normal operation. Giving high priority to the rebuild process, however, can increase response times for incoming application requests which complete for disk service. A balance must be found between acceptable application response times and disk rebuild times. Simulation was used to evaluate the effect of the rebuild unit size on response time and rebuild time. The authors have found this tradeoff to be embodied in the choice of the rebuild unit and the amount of rebuild data which is atomically read from each surviving disk. The find that a single track rebuild unit provides faster rebuild times than a one sector rebuild unit. Rebuilding one track at a time provides better application request response times when compared with rebuilding one cylinder at a time. >

Platelet response to surface modified (SMA) Cobe Duo membrane oxygenators was evaluated in a pig model of cardiopulmonary bypass (CPB) by gamma scintigraphic imaging and platelet impedance aggregometry (PIA). Comparisons of platelet deposition, blood pool platelet counts and platelet aggregation were made between SMA coated CPB circuits and uncoated circuits. The SMA-coated oxygenators exhibited decreases in foreign surface induced platelet aggregation and platelet deposition.

We describe an electrically tunable impedance surface, and antennas based on such a surface. The structure is built as a periodic array of electrically small, coupled resonators, commonly known as a high-impedance surface. By tuning the resonance frequency of the surface as a function of position, one can generate an arbitrary refection phase that varies across the surface. This can be used to steer a reflected radio frequency beam, or to tune antennas that are placed near the surface. We describe such a surface that is electrically tunable using commercially available varactor diodes, and offer techniques for achieving full two-dimensional beam steering with a design that is optimized for bandwidth.

We describe a new kind of beam steering reflector consisting of a textured surface covered with an array of tiny resonators. The surface is mechanically tuned using a motion of only 0.01 wavelength. By individually tuning the resonators, the effective impedance of the surface, and thus the microwave reflection phase, can be varied. By configuring the surface with a reflection phase gradient, it can steer a reflected beam by as much as +/-38°. The concept is also amenable to electrically tunable structures, which can provide faster steering, and greater functionality.

For many years, processor cycle times have continued to increase at a very rapid rate. On top of this, advances in multiprocessor technology have allowed potential system performance to increase at an even faster rate. The result is that the performance of many of today's computer systems is limited by the I/O subsystem. In this paper, the authors attempt to do two things: (1) separate I/O space into three categories, based on their very different raisons d'etre and consequently very different characteristics, and (2) focus on the issues pertaining to improving the performance of one basic mechanism in the I/O subsystem, the magnetic disk. They do the former in order to set the framework of the I/O space. (If one is to improve the performance of the I/O subsystem, one, first has to understand the nature of I/O and not cloud one's efforts by treating I/O as one homogeneous structure.) They do the latter as the first step in dealing with the various mechanisms that make up the I/O space. >