Search for: All records

Mobile Augmented Reality (AR) offers a powerful way to provide spatially-aware guidance for real-world applications. In many cases, these applications involve the configuration of a camera or articulated subject, asking users to navigate several spatial degrees of freedom (DOF) at once. Most guidance for such tasks relies on decomposing available DOF into subspaces that can be more easily mapped to simple 1D or 2D visualizations. Unfortunately, different factorizations of the same motion often map to very different visual feedback, and finding the factorization that best matches a user’s intuition can be difficult. We propose an interactive approach that infers rotational degrees of freedom from short user demonstrations. Users select one or two DOFs at a time by demonstrating a small range of motion, which we use to learn a rotational frame that best aligns with user control of the object. We show that deriving visual feedback from this inferred learned rotational frame leads to improved task completion times on 6DOF guidance tasks compared to standard default reference frames used in most mixed reality applications. 

Our bodies are constantly in motion—from the bending of arms and legs to the less conscious movement of breathing, our precise shape and location change constantly. This can make subtler developments (e.g., the growth of hair, or the healing of a wound) difficult to observe. Our work focuses on helping users record and visualize this type of subtle, longer-term change. We present a mobile tool that combines custom 3D tracking with interactive visual feedback and computational imaging to capture personal time-lapse, which approximates longer-term video of the subject (typically, part of the capturing user’s body) under a fixed viewpoint, body pose, and lighting condition. These personal time-lapses offer a powerful and detailed way to track visual changes of the subject over time. We begin with a formative study that examines what makes personal time-lapse so difficult to capture. Building on our findings, we motivate the design of our capture tool, evaluate this design with users, and demonstrate its effectiveness in a variety of challenging examples. 

Abstract Anomaly, or out-of-distribution, detection is a promising tool for aiding discoveries of new particles or processes in particle physics. In this work, we identify and address two overlooked opportunities to improve anomaly detection (AD) for high-energy physics. First, rather than train a generative model on the single most dominant background process, we build detection algorithms using representation learning from multiple background types, thus taking advantage of more information to improve estimation of what is relevant for detection. Second, we generalize decorrelation to the multi-background setting, thus directly enforcing a more complete definition of robustness for AD. We demonstrate the benefit of the proposed robust multi-background AD algorithms on a high-dimensional dataset of particle decays at the Large Hadron Collider. 

Abstract We study the performance of a cloud-based GPU-accelerated inference server to speed up event reconstruction in neutrino data batch jobs. Using detector data from the ProtoDUNE experiment and employing the standard DUNE grid job submission tools, we attempt to reprocess the data by running several thousand concurrent grid jobs, a rate we expect to be typical of current and future neutrino physics experiments. We process most of the dataset with the GPU version of our processing algorithm and the remainder with the CPU version for timing comparisons. We find that a 100-GPU cloud-based server is able to easily meet the processing demand, and that using the GPU version of the event processing algorithm is two times faster than processing these data with the CPU version when comparing to the newest CPUs in our sample. The amount of data transferred to the inference server during the GPU runs can overwhelm even the highest-bandwidth network switches, however, unless care is taken to observe network facility limits or otherwise distribute the jobs to multiple sites. We discuss the lessons learned from this processing campaign and several avenues for future improvements. 

Abstract Highly granular pixel detectors allow for increasingly precise measurements of charged particle tracks. Next-generation detectors require that pixel sizes will be further reduced, leading to unprecedented data rates exceeding those foreseen at the High- Luminosity Large Hadron Collider. Signal processing that handles data incoming at a rate of $O$(40 MHz) and intelligently reduces the data within the pixelated region of the detectorat ratewill enhance physics performance at high luminosity and enable physics analyses that are not currently possible. Using the shape of charge clusters deposited in an array of small pixels, the physical properties of the traversing particle can be extracted with locally customized neural networks. In this first demonstration, we present a neural network that can be embedded into the on-sensor readout and filter out hits from low momentum tracks, reducing the detector’s data volume by 57.1%–75.7%. The network is designed and simulated as a custom readout integrated circuit with 28 nm CMOS technology and is expected to operate at less than 300  $\mu W$with an area of less than 0.2 mm². The temporal development of charge clusters is investigated to demonstrate possible future performance gains, and there is also a discussion of future algorithmic and technological improvements that could enhance efficiency, data reduction, and power per area.