Damith Senadeera

Damith Senadeera

PhD student at Queen Mary University of London (QMUL); MSc in Data Science, QMUL, 2023; B.Tech in Computer Science and Engineering, NIT-Silchar, India, 2016

Publications

You can also find my articles on my Google Scholar profile.

CUE-Net: Violence Detection Video Analytics with Spatial Cropping Enhanced UniformerV2 and Modified Efficient Additive Attention

Published in IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2024

CUE-Net addresses the challenges in automated violence detection by combining spatial Cropping with an enhanced version of the UniformerV2 architecture, integrating convolutional and self-attention mechanisms alongside a novel Modified Efficient Additive Attention mechanism to effectively and efficiently identify violent activities.

Senadeera, D.C., Yang, X., Kollias, D. and Slabaugh, G., 2024, June. CUE-Net: Violence Detection Video Analytics with Spatial Cropping, Enhanced UniformerV2 and Modified Efficient Additive Attention. In 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

Controlled Text Generation using T5 based Encoder-Decoder Soft Prompt Tuning and Analysis of the Utility of Generated Text in AI

Published in arXiv preprint arXiv:2212.02924, 2022

Investigated and developed a deep learning system to generate synthetic text data using controllable text generation models with the help of prompt tuning given a particular sentiment and analyzed the utility of produced artificial data in AI tasks such as model training.

Senadeera, D.C. and Ive, J., 2022. Controlled text generation using T5 based encoder-decoder soft prompt tuning and analysis of the utility of generated text in AI. arXiv preprint arXiv:2212.02924.