Research Publications

Permanent URI for this communityhttps://rda.sliit.lk/handle/123456789/4194

This main community comprises five sub-communities, each representing the academic contribution made by SLIIT-affiliated personnel.

Browse

Filters

2023 - 20242

Advanced Search

Filter by

Settings

Subject: search.filters.subject.Human computer interaction

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    PublicationOpen Access
    Real-time Multi-spectral Iris Extraction in Diversified Eye Images Utilizing Convolutional Neural Networks
    (IEEE, 2024-07-03) Rathnayake, R; Madhushan, N; Jeeva, A; Darshani, D; Pathirana, I; Ghosh, S; Subasinghe, A; Silva, B N; Wijenayake, U
    Iris extraction has gained prominence due to its application versatility across many domains. However, achieving real-time iris extraction poses challenges due to several factors. Learning-based algorithms outperform non-learning-based iris extraction methods, delivering superior accuracy and performance. In response, this article proposes a Convolutional Neural Networks (CNN)-based, accurate direct iris extraction mechanism for a broad spectrum of eye images. The innovation of our approach lies in its proficiency with varied image types, including those where the iris is partially obscured by the eyelid. We enhance the method’s reliability by introducing a modified Circular Hough Transform (CHT). Extensive testing demonstrates our method’s excellent real-time performance across diverse image types, even under challenging conditions. These findings underscore the proposed method’s potential as a cost-effective and computationally efficient solution for real-time iris extraction in varied application domains.
  • Thumbnail Image
    PublicationEmbargo
    DS-HPE: Deep Set for Head Pose Estimation
    (IEEE, 2023-04-18) Menan, V; Gawesha, A; Samarasinghe, p; Kasthurirathna, D
    Head pose estimation is a critical task that is fundamental to a variety of real-world applications, such as virtual and augmented reality, as well as human behavior analysis. In the past, facial landmark-based methods were the dominant approach to head pose estimation. However, recent research has demonstrated the effectiveness of landmark-free methods, which have achieved state-of-the-art (SOTA) results. In this study, we utilize the Deep Set architecture for the first time in the domain of head pose estimation. Deep Set is a specialized architecture that works on a “set” of data as a result of the “permutation-invariance” operator being utilized in the model. As a result, the model is a simple yet powerful and edge-computation-friendly method for estimating head pose. We evaluate our proposed method on two benchmark data sets, and we compare our method against SOTA methods on a challenging video-based data set. Our results indicate that our proposed method not only achieves comparable accuracy to these SOTA methods but also requires less computational time. Furthermore, the simplicity of our proposed method allows for its deployment in resource-constrained environments without the need for expensive hardware such as Graphics Processing Units (GPUs). This work underscores the importance of accurate and resource-efficient head pose estimation in the fields of computer vision and human-computer interaction, and the Deep Set architecture presents a promising approach to achieving this goal.