Search

Egocentric vision holds great promise for increasing access to visual information and improving the quality of life for blind people. While we strive to improve recognition performance, it remains difficult to identify which object is of interest to the user; the object may not even be included in the frame due to challenges in camera aiming without visual feedback. Also, gaze information, commonly used to infer the area of interest in egocentric vision, is often not dependable. However, blind users tend to include their hand either interacting with the object they wish to recognize or simply placing it in proximity for better camera aiming. We propose a method that leverages the hand as the contextual information for recognizing an object of interest. In our method, the output of a pre-trained hand segmentation model is infused to later convolutional layers of our object recognition network with separate output layers for localization and classification. Using egocentric datasets from sighted and blind individuals, we show that the hand-priming achieves more accurate localization than other approaches that encode hand information. Given only object centers along with labels, our method achieves comparable classification performance to the state-of-the-art method that uses bounding boxes with labels.

A central challenge of social computing research is to enable people to communicate expressively with each other remotely. Augmented reality has great promise for expressive communication since it enables communication beyond texts and photos and towards immersive experiences rendered in recipients' physical environments. Little research, however, has explored AR's potential for everyday interpersonal communication. In this work, we prototype an AR messaging system, ARwand, to understand people's behaviors and perceptions around communicating with friends via AR messaging. We present our findings under four themes observed from a user study with 24 participants, including the types of immersive messages people choose to send to each other, which factors contribute to a sense of immersiveness, and what concerns arise over this new form of messaging. We discuss important implications of our findings on the design of future immersive communication systems., Will be published in PACM HCI, CSCW1, April 2023 issue

This study investigates, numerically, the spatial distribution of metastable helium (He*) in He/SiH4 capacitively coupled plasma (CCP) for the purpose of optimizing plasma density distributions. As a first step, we presented the results of a two-dimensional fluid model of He discharges, followed by those of He/SiH4 discharges to deposit hydrogenated amorphous silicon films, to investigate which factor dominates the coating uniformity. We retained our CCPs in the 300 mm wafer reactor used by the semiconductor industry in the recent past. Selected parameters, such as a sidewall gap (radial distance between the electrode edge and the sidewall), electrical condition of the sidewall, and position of the powered electrode, were considered. In addition, by increasing the gas pressure while varying the sidewall condition, we observed modification of the plasma distributions and, thus, the deposition rate profiles. According to the results, the shift in He* distributions was mainly due to the reduction in the electron mean free path under conditions of gas pressure higher than 100 Pa, as well as local perturbations in the ambipolar electric field due to the finite electrode structure. Small additions of SiH4 largely changed the He* density profile in the midplane of the discharge due to He* quenching. Furthermore, we found that the wide sidewall gap did not improve deposition uniformity against the expectation. This was because the excitation and ionization rate profiles were enhanced and localized only near the bottom electrode edge.

Thermophotovoltaics (TPVs) require emitters with a regulated radiation spectrum tailored to the spectral response of integrated photovoltaic cells. Such spectrally engineered emitters developed thus far are structurally too complicated to be scalable, are thermally unstable, or lack reliability in terms of temperature cycling. Herein, we report wafer-scale, thermal-stress-free, and wavelength-selective emitters that operate for high-temperature TPVs equipped with GaSb photovoltaic cells. One inch crystalline ceria wafers were prepared by sequentially pressing and annealing the pellets of ceria nanoparticles. The direct pyrolysis of citric acid mixed with ceria nanoparticles created agglomerated, pyrolytic carbon and ceria microscale dots, thus forming a carbonized film strongly adhering to a wafer surface. Depending on the thickness of the carbonized film that was readily tuned based on the amount of citric acid used in the reaction, the carbonized ceria emitter behaved as a tungsten-like emitter, a graphite-like emitter, or their hybrid in terms of the absorptivity spectrum. A properly synthesized carbonized ceria emitter produced a power density of 0.63 W/cm2 from the TPV system working at 900 °C, providing 13 and 9% enhancements compared to tungsten and graphite emitters, respectively. Furthermore, only the carbonized ceria emitter preserved its pristine absorptivity spectrum after a 48 h heating test at 1000 °C. The scalable and facile fabrication of thermostable emitters with a structured spectrum will prompt the emergence of thermal emission-harnessed energy devices.

Here, we systematically investigate optical behaviors of ruthenium (Ru)-coated nanoporous anodic aluminum oxide (AAO) nanoarchitectures in the visible and near-Infrared (NIR) regions, where they are fabricated by an atomic layer deposition (ALD) process. Depending on the Ru thickness and the AAO dimensions, the Ru nanoarchitectures could be completely changed (tubular-type or wire-type Ru nanostructures), and the corresponding optical properties are critically affected. Brilliant structural colors are found from the Ru-coated AAO nanostructures, in which the color display covers the full visible range. Beyond the visible region, we also examine the optical behaviors of Ru-coated AAO nanostructures in the NIR region.

Biomass, as harvested, is composed of inorganic compounds both intrinsically and extrinsically and can be quite abrasive. The present study investigates the wear modes and mechanisms of two types of blades of hammer mills used in biomass size reduction (impacting particle size and distribution) and densification (impacting size, shape, and density). The dominant wear modes for the stage 1 steel blades are determined to be erosive and polishing wear. For the stage 2 blades with a carbide weld overlay, the main wear mechanisms are erosion and fracture. Partial replacement of Co by Fe in the outer layer of carbide grits, likely induced by diffusion during the high-temperature welding, has been correlated to the microcracking observed. The microcracking is believed to weaken the grit strength and fracture toughness and make the overlay prone to fracture and erosion due to repetitive contact with the inorganic contents in chopping biomass.

The inorganic content of biomass impairs size reduction tool life and the conversion process. Conventional ash extraction relies on furnace combustion that inevitably alters the inorganic compounds due to oxidation and decomposition. This study developed composition-preserving methods for extracting and analyzing extrinsic and intrinsic inorganic compounds. Comprehensive characterization was carried out on selected biomass feedstocks, including corn stover, pine residue, and pine anatomical fractions, to reveal their inorganic species and morphology. The extrinsic inorganic compounds were found to be dominated by quartz, along with other minor minerals, such as albite, microcline, and gehlenite, and have particle sizes ranging from tens to hundreds of micrometers. Among the pine anatomical fractions, the needles contain the highest intrinsic silicon content while the bark trapped the most extrinsic minerals. By correlation of the total ash and extrinsic inorganic contents to the wear behavior, both the extrinsic and intrinsic inorganic compounds were concluded to have made significant contributions to the wear process. The results here validated a new approach to characterize inorganic compounds in biomass and provided fundamental insights for their potential impact on preprocessing tool wear.

Researchers have adopted remote methods, such as online surveys and video conferencing, to overcome challenges in conducting in-person usability testing, such as participation, user representation, and safety. However, remote user evaluation on hardware testbeds is limited, especially for blind participants, as such methods restrict access to observations of user interactions. We employ smart glasses in usability testing with blind people and share our lessons from a case study conducted in blind participants' homes (N=12), where the experimenter can access participants' activities via dual video conferencing: a third-person view via a laptop camera and a first-person view via smart glasses worn by the participant. We show that smart glasses hold potential for observing participants' interactions with smartphone testbeds remotely; on average 58.7% of the interactions were fully captured via the first-person view compared to 3.7% via the third-person. However, this gain is not uniform across participants as it is susceptible to head movements orienting the ear towards a sound source, which highlights the need for a more inclusive camera form factor. We also share our lessons learned when it comes to dealing with lack of screen reader support in smart glasses, a rapidly draining battery, and Internet connectivity in remote studies with blind participants., Comment: to be published in the proceedings of the 19th International Web for All Conference (Web4All 2022)

The spatial behavior of passersby can be critical to blind individuals to initiate interactions, preserve personal space, or practice social distancing during a pandemic. Among other use cases, wearable cameras employing computer vision can be used to extract proxemic signals of others and thus increase access to the spatial behavior of passersby for blind people. Analyzing data collected in a study with blind (N=10) and sighted (N=40) participants, we explore: (i) visual information on approaching passersby captured by a head-worn camera; (ii) pedestrian detection algorithms for extracting proxemic signals such as passerby presence, relative position, distance, and head pose; and (iii) opportunities and limitations of using wearable cameras for helping blind people access proxemics related to nearby people. Our observations and findings provide insights into dyadic behaviors for assistive pedestrian detection and lead to implications for the design of future head-worn cameras and interactions., Comment: To be published in the proceedings of the 23rd International ACM SIGACCESS Conference on Computers and Accessibility

For people with visual impairments, photography is essential in identifying objects through remote sighted help and image recognition apps. This is especially the case for teachable object recognizers, where recognition models are trained on user’s photos. Here, we propose real-time feedback for communicating the location of an object of interest in the camera frame. Our audio-haptic feedback is powered by a deep learning model that estimates the object center location based on its proximity to the user’s hand. To evaluate our approach, we conducted a user study in the lab, where participants with visual impairments (N = 9) used our feedback to train and test their object recognizer in vanilla and cluttered environments. We found that very few photos did not include the object (2% in the vanilla and 8% in the cluttered) and the recognition performance was promising even for participants with no prior camera experience. Participants tended to trust the feedback even though they know it can be wrong. Our cluster analysis indicates that better feedback is associated with photos that include the entire object. Our results provide insights into factors that can degrade feedback and recognition performance in teachable interfaces.

Egocentric vision holds great promises for increasing access to visual information and improving the quality of life for people with visual impairments, with object recognition being one of the daily challenges for this population. While we strive to improve recognition performance, it remains difficult to identify which object is of interest to the user; the object may not even be included in the frame due to challenges in camera aiming without visual feedback. Also, gaze information, commonly used to infer the area of interest in egocentric vision, is often not dependable. However, blind users often tend to include their hand either interacting with the object that they wish to recognize or simply placing it in proximity for better camera aiming. We propose localization models that leverage the presence of the hand as the contextual information for priming the center area of the object of interest. In our approach, hand segmentation is fed to either the entire localization network or its last convolutional layers. Using egocentric datasets from sighted and blind individuals, we show that the hand-priming achieves higher precision than other approaches, such as fine-tuning, multi-class, and multi-task learning, which also encode hand-object interactions in localization., Comment: the 2020 Winter Conference on Applications of Computer Vision (WACV 2020)

Teachable interfaces can empower end-users to attune machine learning systems to their idiosyncratic characteristics and environment by explicitly providing pertinent training examples. While facilitating control, their effectiveness can be hindered by the lack of expertise or misconceptions. We investigate how users may conceptualize, experience, and reflect on their engagement in machine teaching by deploying a mobile teachable testbed in Amazon Mechanical Turk. Using a performance-based payment scheme, Mechanical Turkers (N = 100) are called to train, test, and re-train a robust recognition model in real-time with a few snapshots taken in their environment. We find that participants incorporate diversity in their examples drawing from parallels to how humans recognize objects independent of size, viewpoint, location, and illumination. Many of their misconceptions relate to consistency and model capabilities for reasoning. With limited variation and edge cases in testing, the majority of them do not change strategies on a second training attempt., Comment: 10 pages, 8 figures, 5 tables, CHI2020 conference

Superhydrophobic (SHPo) surfaces can provide high condensation heat transfer due to facilitated droplet removal. However, such high performance has been limited to low supersaturation conditions due to surface flooding. Here, we quantify flooding resistance defined as the rate of increase in the fraction of water-filled cavities with respect to the supersaturation level. Based on the quantitative understanding of surface flooding, we suggest effective anti-flooding strategies through tailoring the nanoscale coating heterogeneity and structure length scale. Experimental verification is conducted using CuO nanostructures having different length scales combined with hydrophobic coatings with different nanoscale heterogeneities. The proposed anti-flooding SHPo can provide a ∼130% enhanced average heat transfer coefficient with ∼14% larger supersaturation range for droplet jumping compared to a previous CuO SHPo. The proposed anti-flooding parameter and the scalable SHPo will help develop high-performance condensers for real-world applications operating in a wide range of supersaturation levels.

Here, we systematically investigate optical behaviors of ruthenium (Ru)-coated nanoporous anodic aluminum oxide (AAO) nanoarchitectures in the visible and NIR regions, where they are fabricated by using an atomic layer deposition (ALD) process. Depending on the Ru thickness and the AAO dimensions, the Ru nanoarchitectures could be completely changed (tubular-type or wire-type Ru nanostructures), and the corresponding optical properties are critically affected. Brilliant structural colors are found from the Ru-coated AAO nanostructures, in which the color display covers the full visible range. Beyond the visible region, we also examine the optical behaviors of Ru-coated AAO nanostructures in the NIR region.

Blind people have limited access to information about their surroundings, which is important for ensuring one's safety, managing social interactions, and identifying approaching pedestrians. With advances in computer vision, wearable cameras can provide equitable access to such information. However, the always-on nature of these assistive technologies poses privacy concerns for parties that may get recorded. We explore this tension from both perspectives, those of sighted passersby and blind users, taking into account camera visibility, in-person versus remote experience, and extracted visual information. We conduct two studies: an online survey with MTurkers (N=206) and an in-person experience study between pairs of blind (N=10) and sighted (N=40) participants, where blind participants wear a working prototype for pedestrian detection and pass by sighted participants. Our results suggest that both of the perspectives of users and bystanders and the several factors mentioned above need to be carefully considered to mitigate potential social tensions., Comment: The 2020 ACM CHI Conference on Human Factors in Computing Systems (CHI 2020)

Long-Term Evolution-Advanced (LTE-A) networks have been regarded as having great potential to support the Internet of Things (IoT). However, the high probability of preamble collisions when massive numbers of IoT devices try to access the network within a short period has become a significant problem. In this paper, we propose a new scheme to efficiently handle the initial access contention for massive numbers of IoT devices in LTE-A networks. The proposed scheme divides an entire preamble set into $k$ subgroups; then, contention resolution is performed in each subgroup independently in a parallel and pipelined manner. Simulation results show that our scheme leads to considerable improvement in terms of average access delay and standard deviation of access delays without increasing blocking probability. Moreover, utilization of preamble resources is greatly improved by reducing the number of random access slots, where only a few devices participate in preamble selection.

Tribofilms play a vital role in protecting lubricated surfaces in mechanical systems in motion. To date, understanding tribofilms has been mostly based on ex situ analysis. This research investigat...