As 3D Gaussian Splatting (3DGS) emerges as a leading approach for novel view synthesis and scene reconstruction, its potential in digital asset creation has gained significant attention. An increasing number of asset libraries based on GS are being established. However, generating physics-based dynamic assets remains a time-consuming and expertise-intensive task, especially for non-experts. In this paper, we propose LIVE-GS, a highly realistic Virtual Reality (VR) system powered by Large Language Models (LLMs), which enables rapid creation of dynamic Gaussian assets and real-time VR interactions. To inform our system design, we conducted interviews to examine challenges faced by current GS-based VR systems and the specific demands of users. Based on these insights, we employed GPT-4o to analyze key physical properties of objects that significantly impact user interactions, ensuring physics-based interactions in VR align with real-world phenomena. A key innovation of LIVE-GS is its ability to predict reasonable parameters in just 10 seconds from static Gaussian assets while maintaining high-quality VR interactions. To validate our approach, we invited participants experienced in physical simulation to manually adjust physical parameters, providing a baseline for comparison in both asset quality and authoring efficiency. We also conducted a comprehensive user study to evaluate system usability and user satisfaction. Experimental results demonstrate that LIVE-GS, leveraging LLMs' scene understanding capabilities, can achieve efficient physical scene creation and natural interactions without requiring manual design or annotation.

This paper explores finger pressure as a continuous implicit input modality to enhance interaction precision in virtual reality (VR). While motion controllers are widely adopted, their limitations in delicate operations remain a critical challenge. We investigate whether finger pressure signals from conventional VR controllers could offer advantages over traditional kinematic metrics for precision interaction.Through empirical studies, we demonstrate a robust relationship between pressure dynamics and task precision requirements, leading to a lightweight sigmoid-based model that leverages detected pressure to infer desired control granularity. In a comparative evaluation of video-scrubbing tasks, our adaptive method outperforms static sensitivity baselines in both task performance and subjective preference, without elevating cognitive load. Further validation via a VR sketching application demonstrates that our technique maintains task performance while reducing mental demand compared to manual control. Our findings reveal the untapped potential of pressure-based input to bridge coarse and fine-grained VR interactions, offering a path toward more versatile and intuitive input systems.

Abstract As virtual reality display technologies advance, resolutions and refresh rates continue to approach human perceptual limits, presenting a challenge for real‐time rendering algorithms. Neural super‐resolution is promising in reducing the computation cost and boosting the visual experience by scaling up low‐resolution renderings. However, the added workload of running neural networks cannot be neglected. In this article, we try to alleviate the burden by exploiting the foveated nature of the human visual system, in a way that we upscale the coarse input in a heterogeneous manner instead of uniform super‐resolution according to the visual acuity decreasing rapidly from the focal point to the periphery. With the help of dynamic and geometric information (i.e., pixel‐wise motion vectors, depth, and camera transformation) available inherently in the real‐time rendering content, we propose a neural accumulator to effectively aggregate the amortizedly rendered low‐resolution visual information from frame to frame recurrently. By leveraging a partition‐assemble scheme, we use a neural super‐resolution module to upsample the low‐resolution image tiles to different qualities according to their perceptual importance and reconstruct the final output adaptively. Perceptually high‐fidelity foveated high‐resolution frames are generated in real‐time, surpassing the quality of other foveated super‐resolution methods.