T
Thanuja Dharmasiri
Researcher at Monash University
Publications - 17
Citations - 416
Thanuja Dharmasiri is an academic researcher from Monash University. The author has contributed to research in topics: Convolutional neural network & Artificial neural network. The author has an hindex of 10, co-authored 17 publications receiving 325 citations.
Papers
More filters
Proceedings ArticleDOI
Real-Time Joint Semantic Segmentation and Depth Estimation Using Asymmetric Annotations
TL;DR: In this paper, the authors proposed a real-time semantic segmentation network, making changes to further reduce the number of floating point operations, which can handle asymmetric datasets with uneven numbers of annotations per each modality.
Journal ArticleDOI
Telehealth to improve asthma control in pregnancy: A randomized controlled trial
Elida Zairina,Elida Zairina,Michael J. Abramson,Michael J. Abramson,Christine F McDonald,Jonathan C. Li,Thanuja Dharmasiri,Kay Stewart,Susan P. Walker,Susan P. Walker,Eldho Paul,Eldho Paul,Johnson George +12 more
TL;DR: The efficacy of a telehealth programme supported by a handheld respiratory device in improving asthma control during pregnancy in pregnant women is evaluated.
Posted Content
Real-Time Joint Semantic Segmentation and Depth Estimation Using Asymmetric Annotations
TL;DR: This work adapts a recently proposed real-time semantic segmentation network, making changes to further reduce the number of floating point operations, and incorporates the raw predictions of the network into the SemanticFusion framework for dense 3D semantic reconstruction of the scene.
Posted Content
Joint Prediction of Depths, Normals and Surface Curvature from RGB Images using CNNs
TL;DR: In this article, a deep learning-based framework was proposed to estimate depth, surface normals and surface curvature from a single RGB image using a machine learning approach, achieving state-of-the-art performance.
Proceedings ArticleDOI
Just-in-Time Reconstruction: Inpainting Sparse Maps Using Single View Depth Predictors as Priors
TL;DR: This work adopts a fairly standard approach to data fusion, to produce a fused depth map by performing inference over a novel fully-connected Conditional Random Field (CRF) which is parameterized by the input depth maps and their pixel-wise confidence weights.