Facial Expression Recognition Using Transfer Learning with ResNet50

Hiremath, Shantala S.; Hiremath, Jayaprada; Kulkarni, Vaishali; Harshit, Bansal; Kumar, Sujith; Hiremath, Mrutyunjaya S.

doi:10.1007/978-981-99-1624-5_21

Cited by 4 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We assessed the model's efficacy using several metrics: specificity, accuracy, precision, recall, F1 score, and the receiver operating characteristic (ROC) curve. In (6) through 10 provide the mathematical formulations for specificity, accuracy, precision, recall, and F1 score, respectively [30]. Figure 4 illustrates the confusion matrix, displaying the outcomes of machine learning classification based on various combinations of predicted and actual values.…”

Section: Performance Evaluation Parametersmentioning

confidence: 99%

Activity recognition based on spatio-temporal features with transfer learning

Hiremath

et al. 2024

IJ-AI

View full text Add to dashboard Cite

<span lang="EN-US">Human action recognition (HAR) has gathered significant attention recently due to its high demand in various application domains. We explore different CNN architectures for extracting Spatio-Temporal features. In addition to spatial correlation in 2D images, video actions also own the temporal domain attributes. Many research attempts addressed simple activity recognition, but complex activity recognition is still challenging. In the proposed work, the concept of transfer learning consists of two pre-trained networks, ResNet50 and VGG16, which are used to extract Spatiotemporal features and are later fused using the averaging method. Pre-processed frames are considered for the pre-trained network, the pre-processing consists of frame extraction, resizing, and contrast enhancement. The fused features are trained using a multiclass non-linear SVM classifier that achieved an accuracy of 89.71% on the UCF-101 database.</span>

show abstract

Section: Performance Evaluation Parametersmentioning

confidence: 99%

Activity recognition based on spatio-temporal features with transfer learning

Hiremath

et al. 2024

IJ-AI

View full text Add to dashboard Cite

show abstract

“…For example, the classification of cancer on pathological images of cervical tissues using deep learning algorithms was considered by Pan Huang, et al in [6]. The authors used ResNet50 [7], DenseNet12, Inception_v3 [8], and VGGNet19 [9] for an almost balanced dataset consisting of 468 RGB images, including 150 images of the norm, 85 low-grade squamous intraepithelial lesions images, and 104 images of high-grade squamous intraepithelial lesions. Swaraj and Verma [10] solved the COVID-19 classification problem on chest X-ray images.…”

Section: Introductionmentioning

confidence: 99%

Sensitivity of Modern Deep Learning Neural Networks to Unbalanced Datasets in Multiclass Classification Problems

et al. 2023

View full text Add to dashboard Cite

One of the critical problems in multiclass classification tasks is the imbalance of the dataset. This is especially true when using contemporary pre-trained neural networks, where the last layers of the neural network are retrained. Therefore, large datasets with highly unbalanced classes are not good for models’ training since the use of such a dataset leads to overfitting and, accordingly, poor metrics on test and validation datasets. In this paper, the sensitivity to a dataset imbalance of Xception, ViT-384, ViT-224, VGG19, ResNet34, ResNet50, ResNet101, Inception_v3, DenseNet201, DenseNet161, DeIT was studied using a highly imbalanced dataset of 20,971 images sorted into 7 classes. It is shown that the best metrics were obtained when using a cropped dataset with augmentation of missing images in classes up to 15% of the initial number. So, the metrics can be increased by 2–6% compared to the metrics of the models on the initial unbalanced data set. Moreover, the metrics of the rare classes’ classification also improved significantly–the True Positive value can be increased by 0.3 or more. As a result, the best approach to train considered networks on an initially unbalanced dataset was formulated.

show abstract