Domain adaptation for video action recognition by Wang, Xiyu

“…Consequently, this thesis aims to eliminate these assumptions to broaden the applicability of video adaptation methods, focusing on two major shortcomings of conventional VUDA methods, e.g., partial domain adaptation (adapting from a source domain with many classes to a target domain with fewer classes) and continual domain adaptation (adapting to continuously changing target domains). …”
Get full text

Unsupervised domain adaptation for LiDAR segmentation by Kong, Lingdong

“…Transferring knowledge learned from the labeled source domain to the raw target domain for unsupervised domain adaptation (UDA) is essential to the scalable deployment of an autonomous driving system. …”
Get full text

Adaptation of language models via text augmentation by Prachaseree, Chaiyasait

“…This research aims to adapt a language model to a specific domain using text augmentation techniques. …”
Get full text

An adaptive self-biased class-e power amplifier by Min, Ke

“…The Class E power amplifier includes a new adaptive self-biased circuit, which raises the average gate voltage of the common-gate transistor through the fast-charging branch. …”
Get full text

Audio intelligence & domain adaptation for deep learning models at the edge by Ng, Linus JunJia

“…The proposed method and future research in this work will enhance model robustness in adapting to new deployed environments and minimize the manpower time required to acquire and annotate audio data.…”
Get full text

Time series domain adaptation via contrastive adversarial domain disentangled network by Huang, Xinyi

“…A typical unsupervised domain adaptation method is typically designed base on visual data. …”
Get full text

Fake review detection by fusing parameter efficient adapters in pre-trained language model by Ho, See Cheng

“…In addition, multiple adapters can be fused together using the AdapterFusion methodology, opening additional solutions to introduce useful external knowledge into the model. …”
Get full text

The cultural capital of Shakespeare's Hamlet in Haider by Abigail Shantini K.

“…This project explores the concept of cultural capital in relation to Haider, a Hindi film adaptation of Shakespeare's Hamlet. It attempts to understand and exhibit Hamlet's symbolic resonance and show how adaptations of Hamlet add to the source text, bringing previously underemphasized parts of the work to the forefront. …”
Get full text

A cinematic reconsideration of Malay identity through a slice of Hikayat film culture by Muhammad Zhafri Abdul Rahman

“…Hang Jebat is a film adaptation that refocuses the narrative of Hikayat Hang Tuah to depict Hang Jebat, who is traditionally understood as the antagonist, in a more sympathetic light. …”
Get full text

The divorce penalty: an interpretative phenomenological analysis to understand the lived experiences of adult children of divorce in Singapore by Valalan-Rajendran, Sinchita

“…This study found 4 key findings namely stigmatising attitudes towards children from divorced families, parentification, relationship and marital confidence as well as stability as a resilience adaptation. Much research has been aimed at evaluating the effect of divorce on children. …”
Get full text

Investigating the influence of zinc oxide nanoparticles on the pathogenesis of lung fibrosis by Lee, Melissa Kao Hui

“…To date, there have been numerous studies done on the direct ZnO NP-induced cytotoxic response, which exert a variety of cellular effects such as adaptation, oxidative stress, genotoxicity, intracellular organelle dysfunction and apoptosis. …”
Get full text

Deep neural network compression for pixel-level vision tasks by He, Wei

“…Finally, we propose Adaptive Correlation-driven Sparsity Learning (ACSL) for DCNN compression that can provide superior performance on both image-level and pixel-level vision tasks. …”
Get full text

Object counting using machine learning by Tang, Haihan

“…In this thesis, to improve the accuracy of multi-modal crowd count estimation, a three-stream adaptive fusion network (TAFNet) and a scale-aware self-differential attention network (SDANet) are proposed. …”
Get full text

Automated acne assessment on selfie by Li, Yihang

“…Transfer learning skills are also used to improve the performance of the adapted meta learning network on CauAcne. After training, the assessment accuracy reached 64% and outperformed the state of the art methods by 30%.…”
Get full text

Transient disabilities: didactic illness, corrective pain and enforced care relations in the “Golden Age” of children's literature by Khaira, Tasreen Kaur

“…Paying special attention to texts that continue to be reprinted and adapted today—such as Frances Hodgson Burnett’s The Secret Garden and its film adaptations—this thesis will trace the extent to which they continue to perpetuate hegemonic structures of able-bodied normativity and femininity despite the shifting social and cultural attitudes towards disability today.…”
Get full text

Cinematic virtual reality film practice: expanded profilmic event and mode of production by Chan, Fong Yuen

“…While the long-established conventional film practices can merely be directly adapted to the expanded profilmic event on CVR film sets, the job responsibilities of CVR major creative roles are no longer the same as conventional roles. …”
Get full text

Multi-camera tracking for smart urban mobility by Do, Van Minh

“…We first introduce the algorithmic augmentations that facilitate adaptive region-of-interest (ROI) to restrict computations only on relevant scene regions, along with the use of sparse optical flow and linear estimation to improve tracking accuracy. …”
Get full text

When “We” were there: comprehension of Singapore's historic locations through audience engagement in VR by Fang, Dexter Zong Rui

“…The research intends to adapt VR’s aspects and interaction possibilities to present these locations. …”
Get full text

Effective action recognition with fully supervised and self-supervised methods by Cao, Haozhi

“…Key points are adaptively extracted as feature points with maximum feature values at split regions, while key point shifts are the spatial displacements of corresponding key points. …”
Get full text

Real-time object tracking for event cameras by Zhang, Yexin

“…Event data are modeled as spacetime event clouds and fed to an adapted PointNet architecture to extract spatial and temporal information of the target object. …”
Get full text