CommuSpotter: Scene Text Spotting with Multi-Task Communication
Scene text spotting is a challenging multi-task modulation for locating and recognizing texts in complex scenes. Existing end-to-end text spotters generally adopt sequentially decoupled multi-tasks, consisting of text detection and text recognition ...
Liang Zhao, Greg Wilsbacher, Song Wang
doaj +1 more source
On Calibration of Scene-Text Recognition Models
In this work, we study the problem of word-level confidence calibration for scene-text recognition (STR). Although the topic of confidence calibration has been an active research area for the last several decades, the case of structured and sequence prediction calibration has been scarcely explored.
Ron Slossberg +8 more
openaire +2 more sources
Implementation of a Human-Computer Interface for Computer Assisted Translation and Handwritten Text Recognition [PDF]
A human-computer interface is developed to provide services of computer assisted machine translation (CAT) and computer assisted transcription of handwritten text images (CATTI). The back-end machine translation (MT) and handwritten text recognition (HTR)
Ocampo Sepúlveda, Jorge Carlos
core +1 more source
Efficient Invariant Features for Sensor Variability Compensation in Speaker Recognition [PDF]
In this paper, we investigate the use of invariant features for speaker recognition. Owing to their characteristics, these features are introduced to cope with the difficult and challenging problem of sensor variability and the source of performance ...
Abderrezak Guessoum +5 more
core +1 more source
The ubiquitin ligase RNF115 is required for the clearance of damaged lysosomes
Upon lysosomal rupture, an E3 ubiquitin ligase RNF115 translocates from the cytosol to the damaged lysosomal membrane. Moreover, RNF115 depletion impairs the clearance of damaged lysosomes, identifying it as a key regulator of lysosomal quality control.
Sae Nakanaga +3 more
wiley +1 more source
Research on vectorization technology for electrical wiring diagrams
Electrical wiring diagrams, which represent the positions of electrical equipment and describe their wiring and cabling methods, need to be converted into XML for smart grid system scheduling and management.
ZHANG Yong +5 more
doaj +1 more source
Noise-Robust Wagon Text Extraction Based on Defect-Restore Generative Adversarial Network
Wagon text extraction mainly depends on manual identification of relevant information, which is laborious, time consuming, monotonous and error-prone. To address this concern, we develop a two-stage wagon text extraction system based on the combination ...
Meng Lei +5 more
doaj +1 more source
Random Blur Data Augmentation for Scene Text Recognition
In this paper, we propose to apply data augmentation approaches that provide more diverse training images, thus helping train more robust deep models for the Scene Text Recognition (STR) task.
Deguo Mu, Wei Sun, Guoliang Xu, Wei Li
doaj +1 more source
The ubiquitin‐proteasome system and autophagy as guardians of the cellular proteome
This Perspective covers the three principles governing the crosstalk between the ubiquitin‐proteasome system and autophagy in cellular proteostasis: (1) a shared ubiquitin code routing substrates via shuttle factors or autophagy receptors; (2) spatial compartmentalization into phase‐separated degradation hubs and organelle‐specific modules (exemplified
Ivan Dikic
wiley +1 more source
A Remote Calibration Device Using Edge Intelligence
Power system facility calibration is a compulsory task that requires in-site operations. In this work, we propose a remote calibration device that incorporates edge intelligence so that the required calibration can be accomplished with little human ...
Quan Wang +4 more
doaj +1 more source

