English
norskInstitutt for fysikk og teknologi
Bla i:
Enhetens samlinger
Nye registreringer
-
PSAIR: A Neuro-Symbolic Approach to Zero-Shot Visual Grounding
(Chapter; Bokkapittel, 2024-09-09)Supervised methods for Visual Grounding often require costly annotations of paired sentences and images with ground truth boxes. Recent zero-shot approaches to visual grounding such as ReCLIP and ChatRef aim to avoid the need for costly annotation of paired sentences and images with ground truth boxes. However, these approaches leverage an inflexible detect-then-reasoning paradigm, which leads to a ... -
Interrogating Sea Ice Predictability with Gradients
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-02-14)Predicting sea ice concentration (SIC) is an important task in climate analysis. The recently proposed deep learning system IceNet is the state-of-the-art sea ice prediction model. IceNet takes high-dimensional climate simulations and observational data as input features and forecasts SIC for the next 6 months over a spatial grid over the northern hemisphere. The model has proven to be particularly ... -
Tailored multi-layer graphene arrays for precision detection of neurotransmitter
(Journal article; Tidsskriftartikkel, 2024-05-22)The development of sensitive dopamine (DA) sensors has significantly contributed to our understanding of DA-related disorders and has enabled advancements in medical diagnostics and neuroscience research. Therefore, in the present work, a novel electrochemical sensing platform was developed using multi-layer graphene arrays (MGA) modified electrodes for the selective detection of DA. The preparation ... -
Nano-cerium zinc molybdate embedded with activated graphene for detection of levofloxacin in polluted water resources
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-05-28)Antibiotics have widespread applications in personal care, animal muscle growth, and aquaculture, yet their pervasive use gives rise to substantial risks for human health and the ecosystem. Consequently, the imperative lies in developing accurate and highly sensitive detection techniques for analyzing levofloxacin (LFX). This research focuses on the hydrothermal synthesis of unique three-dimensional ... -
Space weather mapped by millions of smartphones
(Journal article; Tidsskriftartikkel, 2024-11-13)Phone users around the world are enabling the creation of a space-weather monitor that will deepen our understanding of the physics governing Earth’s ionized upper atmosphere and improve the accuracy of satellite positioning. -
Compact Linnik-type hyperspectral quantitative phase microscope for advanced classification of cellular components
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-06-20)Hyperspectral quantitative phase microscopy (HS-QPM) involves the acquisition of phase images across narrow spectral bands, which enables wavelength-dependent study of different biological samples. In the present work, a compact Linnik-type HS-QPM system is developed to reduce the instability and complexity associated with conventional HS-QPM techniques. The use of a single objective lens for both ... -
Finding NEM-U: Explaining unsupervised representation learning through neural network generated explanation masks
(Journal article; Tidsskriftartikkel, 2024)Unsupervised representation learning has become an important ingredient of today’s deep learning systems. However, only a few methods exist that explain a learned vector embedding in the sense of providing information about which parts of an input are the most important for its representation. These methods generate the explanation for a given input after the model has been evaluated and tend to ... -
Sculptured silicon nanopillars bridging face to face nanogaps with metal-insulator-metal coating for surface enhanced Raman spectroscopy
(Journal article; Tidsskriftartikkel; Peer reviewed, 2023-12-29)A multilayer structure of metal-insulator-metal (MIM) thin film on a sculptured silicon nanopillar (Si NP) is designed and optimized for surface-enhanced Raman spectroscopy (SERS) applications. A facile fabrication method of large-area periodic nanopillars was developed by combining reactive ion etching and monolayer polystyrene (PS) beads on a Si wafer to fabricate a regular array of Si NPs. The ... -
BrainIB: Interpretable Brain Network-Based Psychiatric Diagnosis With Graph Information Bottleneck
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-09-13)Developing new diagnostic models based on the underlying biological mechanisms rather than subjective symptoms for psychiatric disorders is an emerging consensus. Recently, machine learning (ML)-based classifiers using functional connectivity (FC) for psychiatric disorders and healthy controls (HCs) are developed to identify brain markers. However, existing ML-based diagnostic models are prone to ... -
Improved electrochemical detection of levofloxacin in diverse aquatic samples using 3D flower-like Co@CaPO4 nanospheres
(Journal article; Tidsskriftartikkel; Peer reviewed, 2023-12-18)The misuse of antibiotics has become a concerning environmental issue, posing a significant threat to public health. Levofloxacin (LFX), a fluoroquinolone antibiotic, is particularly worrisome due to its detrimental impact on human health and the ecosystem. Therefore, the selective and accurate identification of LFX is of utmost importance. In this study, we have developed an electrochemical sensor ... -
Experimental demonstration of a dispatchable power-to-power high temperature latent heat storage system
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-03-23)The development of electricity storage solutions is crucial to support the integration of variable renewable electricity sources in electricity systems. This study experimentally characterizes a state-of-the-art full-scale electrical thermal energy storage (ETES) system in outdoor conditions. The system integrates a 600 kWh<sub>th</sub> high-temperature latent heat storage module and a 13 kW<sub>e</sub> ... -
Iceberg Drift Trajectories from Optical Satellite Data in the Barents Sea
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-06)Observed iceberg drift trajectories are needed to understand iceberg drift, and to validate ice- berg drift and deterioration models. Herein, drift trajectories of small and medium icebergs are derived from a sequence of optical Sentinel-2, Landsat, and Planet satellite acquisitions over eight days in July 2022 at Severny Island in the Barents Sea. In total, 246 icebergs are tracked visually in ... -
Deep learning-based postoperative glioblastoma segmentation and extent of resection evaluation: Development, external validation, and model comparison
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-11-16)Background - The pursuit of automated methods to assess the extent of resection (EOR) in glioblastomas is challenging, requiring precise measurement of residual tumor volume. Many algorithms focus on preoperative scans, making them unsuitable for postoperative studies. Our objective was to develop a deep learning-based model for postoperative segmentation using magnetic resonance imaging (MRI). We ... -
Single-shot quantitative differential phase contrast microscope using a single calcite beam displacer
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-11-06)This paper presents the development of a single-shot, partially spatially coherent quantitative differential phase contrast microscopy (Q-DPCM) system. The optical scheme comprises a polarizer, lenses, calcite beam displacer, and analyzer, which can be seamlessly integrated to an existing bright-field microscopy system, transforming it into a Q-DPCM system. It utilizes a partially spatially coherent ... -
Diffusion Models with Cross-Modal Data for Super-Resolution of Sentinel-2 To 2.5 Meter Resolution
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-09-05)Diffusion models have obtained photo-realistic results on various super-resolution tasks. However, existing approaches typically require the availability of high-resolution and paired training data, which often is not readily available in many remote sensing scenarios. To enhance multi-spectral Sentinel 2 (S2) satellite images – at a ground sampling distance (GSD) ranging from 10m to 60m – without ... -
PTUS: Photo-Realistic Talking Upper-Body Synthesis via 3D-Aware Motion Decomposition Warping
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-03-24)Talking upper-body synthesis is a promising task due to its versatile potential for video creation and consists of animating the body and face from a source image with the motion from a given driving video. However, prior synthesis approaches fall short in addressing this task and have been either limited to animating heads of a target person only, or have animated the upper body but neglected the ... -
Simultaneous measurement of refraction and absorption with an integrated near-infrared Mach–Zehnder interferometer
(Journal article; Tidsskriftartikkel, 2024-05-22)Most integrated evanescent-field photonic sensors measure changes in either the real part or the imaginary part of the complex refractive index of the sample, i.e., refraction or absorption. Here we propose and experimentally demonstrate a near-infrared sensor based on a silicon nitride Mach-Zehnder interferometer which provides a direct measurement of the complex refractive index. Our architecture ... -
ExMap: Leveraging Explainability Heatmaps for Unsupervised Group Robustness to Spurious Correlations
(Journal article; Tidsskriftartikkel, 2024-09-16)Group robustness strategies aim to mitigate learned biases in deep learning models that arise from spurious correlations present in their training datasets. However, most existing methods rely on the access to the label distribution of the groups, which is time-consuming and expensive to obtain. As a result, unsupervised group robustness strategies are sought. Based on the insight that a trained ... -
Visual Data Diagnosis and Debiasing with Concept Graphs
(Journal article; Tidsskriftartikkel, 2024-09-26)The widespread success of deep learning models today is owed to the curation of extensive datasets significant in size and complexity. However, such models frequently pick up inherent biases in the data during the training process, leading to unreliable predictions. Diagnosing and debiasing datasets is thus a necessity to ensure reliable model performance. In this paper, we present ConBias, a novel ... -
Projection of future non-stationary intensity-duration-frequency curves using the pooled CMIP6 climate models
(Journal article; Tidsskriftartikkel; Peer reviewed, 2024-07-13)Extreme precipitation events can cause severe floods that pose significant risks to human lives, properties, and ecosystems. Therefore, understanding how climate change may affect the characteristics of these events is crucial for developing effective adaptation and mitigation strategies. In this study, we investigated the effect of climate change on the extreme characteristics through the concept ...
RSS 1.0