Improved Mixed-Example Data Augmentationdoi:10.1109/WACV.2019.00139Cecilia SummersMichael J. DinneenIEEEWorkshop on Applications of Computer Vision
The input adopts adaptive image scaling and mixed data augmentation techniques to ensure that various input images are scaled to a uniform size, thereby meeting the requirements of the backbone network for input sizes. The backbone network comprises three parts, MobileNetv3, E-ELAN, and MP-1, ...
For example, data augmentation strategy fixes up problems of insufficient information of small target and texture, but increases computing costs. Guo et al.19 introduced a new deep-learning model named Small Target CenterNet, which applies the selective small target replication algorithm (SSTRA)and ...
However, in field conditions or mixed-species habitats, increased variability in facial features and behavioral patterns introduces significant recognition challenges. Future investigations could focus on developing adaptive recognition frameworks capable of accommodating environmental variability while maintaining ...
This study introduces a novel data augmentation method aimed at addressing the limited availability of sample data for strip steel surface defects, utilizing an improved Auxiliary Classifier Generative Adversarial Network (ACGAN). Firstly, on the generator side, the improved ACGAN integrates an encoder ...
For example, Yang et al. [6] proposed a contrastive learning approach for sequential recommendation that uses graph neural networks to guide data augmentation and importance coding, which can improve the quality and diversity of contrast views. Liu et al. [10] presented a dual-knowledge view ...
(2020) mixed CNN and GRU models for time series prediction. Some scholars have combined metaheuristic learning methods with neural networks for time series prediction. For example, Abdulkarim and Engelbrecht (2021) used PSO to optimize feedforward neural networks for time series prediction in dynamic...
As complexity and capabilities of Artificial Intelligence technologies increase, so does its potential for misuse. Deepfake videos are an example. They are
In this work, we explore how the acoustic modeling (AM) can benefit from monolingual speech data belonging to the high-resourced mixed language. For this purpose, we train state-of-the-art AMs, which were ineffective due to lack of training data, on a significantly increased amount of CS ...
python scripts/check_dataset.py [-s --source DATASET_PATH]#[optional] Default: data/SFID_demo Train model To train a model, run: python train.py [-s --source DATASET_PATH]#[optional] Default: data/SFID_demo[-w --weights WEIGHTS_PATH]#[optional] Default: None[--epochs EPOCHS]#[optio...