Cross-Dataset Data Augmentation for Convolutional Neural Networks Training

Gasparetto, Andrea; Ressi, Dalila; Bergamasco, Filippo; Pistellato, Mara; Cosmo, Luca; Boschetti, Marco Antonio; Ursella, Enrico; Albarelli, Andrea

doi:10.1109/icpr.2018.8545812

Cited by 7 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of classification, traditional word representation methods such as TF-IDF (and, to a lesser extent, word embeddings) have been widely utilized as input features for traditional classification methods such as decision trees [49,50], support vector machines [51,52], and probabilistic graphical models (e.g., naive Bayes and hidden Markov models) [53,54]. The same can be said about word embeddings, which have, however, seen much greater use with specialized neural network architectures such as convolutional neural networks (CNNs) [55][56][57][58] and RNNs [59][60][61]. Transformer-based LMs such as the Bidirectional Encoder Representations from Transformers (BERT) [34] and Generative Pre-trained Transformer (GPT) [36], on the other hand, have showcased outstanding classification results by passing the contextualized embeddings through a simple feed-forward layer.…”

Section: Classificationmentioning

confidence: 99%

Hierarchical Text Classification and Its Foundations: A Review of Current Research

Zangari,

Marcuzzo,

Rizzo

et al. 2024

Electronics

Self Cite

View full text Add to dashboard Cite

While collections of documents are often annotated with hierarchically structured concepts, the benefits of these structures are rarely taken into account by classification techniques. Within this context, hierarchical text classification methods are devised to take advantage of the labels’ organization to boost classification performance. In this work, we aim to deliver an updated overview of the current research in this domain. We begin by defining the task and framing it within the broader text classification area, examining important shared concepts such as text representation. Then, we dive into details regarding the specific task, providing a high-level description of its traditional approaches. We then summarize recently proposed methods, highlighting their main contributions. We also provide statistics for the most commonly used datasets and describe the benefits of using evaluation metrics tailored to hierarchical settings. Finally, a selection of recent proposals is benchmarked against non-hierarchical baselines on five public domain-specific datasets. These datasets, along with our code, are made available for future research.

show abstract

Section: Classificationmentioning

confidence: 99%

Hierarchical Text Classification and Its Foundations: A Review of Current Research

Zangari,

Marcuzzo,

Rizzo

et al. 2024

Electronics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Nowadays, learning-based techniques are employed in a variety of applications, being able to solve even complex problems [ 25 , 26 , 27 ]. Researchers have also investigated the creation of HDR using multiple LDR images by designing deep learning methods.…”

Section: Related Workmentioning

confidence: 99%

Exploiting Light Polarization for Deep HDR Imaging from a Single Exposure

Pistellato

Fatima

Wimmer

2023

Sensors

Self Cite

View full text Add to dashboard Cite

In computational photography, high dynamic range (HDR) imaging refers to the family of techniques used to recover a wider range of intensity values compared to the limited range provided by standard sensors. Classical techniques consist of acquiring a scene-varying exposure to compensate for saturated and underexposed regions, followed by a non-linear compression of intensity values called tone mapping. Recently, there has been a growing interest in estimating HDR images from a single exposure. Some methods exploit data-driven models trained to estimate values outside the camera’s visible intensity levels. Others make use of polarimetric cameras to reconstruct HDR information without exposure bracketing. In this paper, we present a novel HDR reconstruction method that employs a single PFA (polarimetric filter array) camera with an additional external polarizer to increase the scene’s dynamic range across the acquired channels and to mimic different exposures. Our contribution consists of a pipeline that effectively combines standard HDR algorithms based on bracketing and data-driven solutions designed to work with polarimetric images. In this regard, we present a novel CNN (convolutional neural network) model that exploits the underlying mosaiced pattern of the PFA in combination with the external polarizer to estimate the original scene properties, and a second model designed to further improve the final tone mapping step. The combination of such techniques enables us to take advantage of the light attenuation given by the filters while producing an accurate reconstruction. We present an extensive experimental section in which we validate the proposed method on both synthetic and real-world datasets specifically acquired for the task. Quantitative and qualitative results show the effectiveness of the approach when compared to state-of-the-art methods. In particular, our technique exhibits a PSNR (peak signal-to-noise ratio) on the whole test set equal to 23 dB, which is 18% better with respect to the second-best alternative.

show abstract

“…This paper mainly discussed about the comparision of accuracy results for augmented and un augmented dataset. In [5] The additional samples are given by mapping elements from a different pool rather than from the dataset itself, according to a new data augmentation approach proposed. Cross-Dataset Data Augmentation was proposed and demonstrated sucessfully.…”

Section: Literature Surveymentioning

confidence: 99%

Data Set Preparation Using the Method of Data Augmentation for Classification of Skin Diseases

J.A¹,

MuraliBhaskharan²

2021

Advances in Parallel Computing

View full text Add to dashboard Cite

Skin disease are mostly ignored and provided less importance at the early stages. Some ignorance among people might lead to skin cancer. In existing approach, the increased skin disease are identified at the later stage using biopsy only. The inspection is performed manually by considering many histopathological features. This paper explains the development of a model which helps to detect the various skin diseases at early stage using neural networks. For classification and prediction purpose this paper using Convolutional Neural Network. For improving the accuracy of prediction we are going to do dataset preparation. Data augmentation is using for increasing the count size of the training data. So that the accuracy of the classification will increase and will get 97% result.

show abstract

Cross-Dataset Data Augmentation for Convolutional Neural Networks Training

Cited by 7 publications

References 9 publications

Hierarchical Text Classification and Its Foundations: A Review of Current Research

Hierarchical Text Classification and Its Foundations: A Review of Current Research

Exploiting Light Polarization for Deep HDR Imaging from a Single Exposure

Data Set Preparation Using the Method of Data Augmentation for Classification of Skin Diseases

Contact Info

Product

Resources

About