Skip to search formSkip to main contentSkip to account menu

A Study on CNN Transfer Learning for Image Classification

This work proposes the study and investigation of such a CNN architecture model (i.e. Inception-v3) to establish whether it would work best in terms of accuracy and efficiency with new image datasets via Transfer Learning, and the results are compared to some state-of-the-art approaches.
View via Publisher (opens in a new tab)

A probabilistic approach for human everyday activities recognition using body motion from RGB-D images

An approach that relies on cues from depth perception from RGB-D images, where features related to human body motion are used on multiple learning classifiers in order to recognize human activities on a benchmark dataset, overcomes state of the art methods in terms of precision, recall and overall accuracy.
View on IEEE (opens in a new tab)

A Study on Mental State Classification using EEG-based Brain-Machine Interface

Results show that only 44 features from a set of over 2100 features are necessary when used with classical classifiers such as Bayesian Networks, Support Vector Machines and Random Forests, attaining an overall accuracy over 87%.
View on IEEE (opens in a new tab)

A Deep Evolutionary Approach to Bioinspired Classifier Optimisation for Brain-Machine Interaction

An evolutionary-optimised MLP achieves results close to the Adaptive Boosted LSTM for the two first experiments and significantly higher for the number-guessing experiment with an Adaptive boosted DEvo MLP reaching 31.35%, while being significantly quicker to train and classify.
View PDF on arXiv (opens in a new tab)

Cross-Domain MLP and CNN Transfer Learning for Biological Signal Processing: EEG and EMG

The success of unsupervised transfer learning between Electroencephalographic (brainwave) classification and Electromyographic (muscular wave) domains with both MLP and CNN methods is shown, indicating that knowledge transfer is possible between the two signal domains.
View on IEEE (opens in a new tab)

Mental Emotional Sentiment Classification with an EEG-based Brain-machine Interface

Of the set of 2548 features, a subset of 63 selected by their Information Gain values were found to be best when used with ensemble classifiers such as Random Forest, achieving an overall accuracy of around 97.89%, outperforming the current state of the art by 2.99 percentage points.

SocNav1: A Dataset to Benchmark and Learn Social Navigation Conventions

SocNav1 aims at evaluating the robots’ ability to assess the level of discomfort that their presence might generate among humans and is particularly well-suited to be used to benchmark non-Euclidean machine learning algorithms such as graph neural networks.
View PDF on arXiv (opens in a new tab)

A human activity recognition framework using max-min features and key poses with differential evolution random forests classifier

View via Publisher (opens in a new tab)

Extracting data from human manipulation of objects towards improving autonomous robotic grasping

View via Publisher (opens in a new tab)

Stepping-stones to Transhumanism: An EMG-controlled Low-cost Prosthetic Hand for Academia

This work intends to be a pioneer into developing a low-cost multipurpose robotic hand for research and academia and contribute to an ever-increasing human-robot symbiosis by motivating students to engage in transhumanism studies using more sophisticated technologies and methods.
View on IEEE (opens in a new tab)
...
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)