Results 31 to 40 of about 539,870 (246)

Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches

open access: yesSensors
Human activity recognition (HAR) using wearable sensors has advanced through various machine learning paradigms, each with inherent trade-offs between performance and labeling requirements.
Taoran Sheng, Manfred Huber
doaj   +1 more source

Weakly Supervised Solar Panel Mapping via Uncertainty Adjusted Label Transition in Aerial Images

open access: yes, 2023
This paper proposes a novel uncertainty-adjusted label transition (UALT) method for weakly supervised solar panel mapping (WS-SPM) in aerial images. In weakly supervised learning (WSL), the noisy nature of pseudo labels (PLs) often leads to poor model ...
Hu, J, Zhou, J, Jia, X, Zhang, J
core   +1 more source

Design and analysis strategies for robust microbiome ageing research

open access: yesFEBS Letters, EarlyView.
The gut microbiome changes with age and associates with age‐related morbidity and mortality, establishing it as a potential biomarker and intervention target for ageing. Realising this potential requires methodological rigour, yet distinguishing biological signals from methodological artefacts remains challenging across cohorts. This review provides an
Mark Olenik   +5 more
wiley   +1 more source

Investigating transcription factor dynamics in health and disease using FRAP

open access: yesFEBS Letters, EarlyView.
FRAP analysis of GFP‐tagged transcription factors reveals how molecular mobility and target engagement change in response to drug treatment. By combining live‐cell imaging, quantitative model fitting, and statistical analysis, this approach uncovers transcription factor dynamics linked to disease mechanisms, providing a powerful framework for ...
Kannan Govindaraj   +3 more
wiley   +1 more source

Weakly Supervised Violence Detection in Surveillance Video

open access: yesSensors, 2022
Automatic violence detection in video surveillance is essential for social and personal security. Monitoring the large number of surveillance cameras used in public and private areas is challenging for human operators.
David Choqueluque-Roman   +1 more
doaj   +1 more source

Golgi enzymes are retrieved from the plasma membrane to the trans‐Golgi network

open access: yesFEBS Letters, EarlyView.
Golgi enzymes are traditionally considered resident proteins retained within the Golgi apparatus. Here, we demonstrate that a subset transiently reaches the cell surface and is subsequently retrieved to the trans‐Golgi network via retrograde transport. Using a nanobody‐based toolkit, we uncover a dynamic trafficking cycle of several Golgi enzymes.
Dominik P. Buser, Tina Junne
wiley   +1 more source

Detecting rare events using Kullback-Leibler divergence: A weakly supervised approach [PDF]

open access: yes, 2016
Video surveillance infrastructure has been widely installed in public places for security purposes. However, live video feeds are typically monitored by human staff, making the detection of important events as they occur difficult.
Denman, Simon   +3 more
core   +1 more source

Structural and biochemical analysis of a B12 superbinder

open access: yesFEBS Letters, EarlyView.
BtuG proteins are vitamin B12 scavengers in Bacteroides thetaiotaomicron, a dominant human gut bacterium. We present crystal structures of three BtuG homologs bound to cobalamin and its precursor cobinamide, revealing picomolar binding affinities, among the highest known for any natural protein.
Jose M. Martinez Felices   +3 more
wiley   +1 more source

Lesion region segmentation via weakly supervised learning

open access: yesQuantitative Biology, 2022
Background Image‐based automatic diagnosis of field diseases can help increase crop yields and is of great importance. However, crop lesion regions tend to be scattered and of varying sizes, this along with substantial intra‐class variation and small ...
Ran Yi   +5 more
doaj   +1 more source

SinkholeNet: A novel RGB-slope sinkhole dataset and deep weakly-supervised learning framework for sinkhole classification and localization

open access: yesEgyptian Journal of Remote Sensing and Space Sciences, 2023
This paper proposes a novel multimodal deep weakly-supervised learning framework, SinkholeNet, to classify and localize sinkhole(s) in high-resolution RGB-slope aerial images.
Amir Yavariabdi   +6 more
doaj   +1 more source

Home - About - Disclaimer - Privacy