Human Action Recognition using Pre-trained Convolutional Neural Networks

Aly, Cherry Aly; Abas, Fazly Salleh; Goh, Hock Ann

doi:10.1145/3442705.3442710

Cited by 1 publication

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A pre-trained VGG16 framework together with transfer learning was selected for classification. Ali et al [34] compared the performance of two pre-trained CNN models with different network architectures, i.e., GoogleNet and AlexNet on human action recognition datasets. Feature vectors are extracted from a video and trained via the transfer learning (TL) paradigm using Long-Short Term Memory (LSTM) framework to predict the video action labels.…”

Section: B Fusion Methodsmentioning

confidence: 99%

Transfer Learning Models for CNN Fusion With Fisher Vector for Codebook Optimization of Foreground Features

Kamaleldin,

Abu-Bakar,

Sheikh

2024

IEEE Access

View full text Add to dashboard Cite

Human action recognition has become one of the main topics in the computer vision field due to its high demand and competitiveness in real-world applications. The main goals of human action recognition are to improve classification accuracy and reduce computational complexity. Previous studies have mainly used two approaches: the handcrafted feature extraction approach and the deep learning approach. The hand-crafted approach is simple, which confers it with an added advantage in terms of computational complexity. However, this method is low in accuracy. Conversely, the deep learning approach achieves high accuracy even for complex datasets, but it suffers in terms of computational complexity and long training time as it needs to process huge datasets during training. Other approaches include the use of pre-trained deep learning networks to fuse both methods. In this paper, we will introduce a combination of pre-trained convolutional neural networks (CNN) to extract features, an improved Fisher vector (iFV) codebook, and an optimized support vector machine SVM to achieve improved human action recognition. We leveraged three pre-trained CNNs, namely, Inception-ResNet-v2, NASNet-Large, and Xception, to extract the features. Then, we applied the improved Fisher vector codebook to encode them. We subsequently trained the codebook using SVM for classification and re-adjusted the SVM weights using five different optimization techniques, which are SGD, Adagrad, ADAM, Adamax, and Nadam. To evaluate the performance, we utilized UCF101 and HMDB51 datasets. The results demonstrate that the accuracy and computational complexity of our approach are comparable to state-of-the-art techniques.

show abstract

Section: B Fusion Methodsmentioning

confidence: 99%