Scenes with water surfaces present a significant challenge for Gaussian Splatting due to the simultaneous presence of refraction and reflection, as well as the difficulty of accurately estimating the geometry of transparent water surfaces. To address this, we propose a novel framework for reconstructing scenes involving both reflection and refraction caused by water surfaces. The water surface is modeled as a trainable plane, and 2D Gaussian ray tracing is applied to account for refraction through the water. We extend 2D Gaussian Splatting by introducing a soft mask parameter and a dual set of Gaussian primitives, which handle both reflected and refracted effects. Our method achieves state-of-the-art performance on newly constructed water surface datasets, including both synthetic and real scenes, and significantly outperforms prior approaches in water-interacting regions. Furthermore, we demonstrate the editability of our model by manipulating the index of refraction to suppress or modify refractive effects, enabling scene transformations into different liquids.

Recent research on LiDAR-based 3D object detectors has shown strong performance; however, evaluations typically focus on dominant classes, overlooking rare classes, such as strollers, which could be critical in real autonomous driving scenarios. This oversight is problematic because state-of-the-art 3D object detectors show significantly lower performance on rare classes compared to dominant ones when trained on both. To address this issue and achieve accurate 3D rare object detection using only LiDAR data, we propose the Neighbor-Based confidence Adjustment for 3D rare class predictions (NBA3D). NBA3D utilizes a graph neural network to analyze the surrounding environment of rare class prediction boxes, enabling a more effective distinction between true positives and false positives based on their local context. Our approach utilizes both 3D prediction box characteristics and CLIP-based class semantic information to better contextualize neighboring objects. Various experiments demonstrate that NBA3D effectively improves the detection performance of rare class objects, regardless of the type of 3D object detectors used.

Face frontalization, which aims to synthesize a frontal view from an arbitrary facial pose, plays a crucial role in enhancing downstream tasks such as face recognition, expression analysis, and lip reading. However, previous methods often rely on paired or annotated datasets, which are costly and impractical to obtain, particularly in unconstrained real-world scenarios. To overcome this limitation, we propose a novel framework that leverages pretrained generative and encoder-based projection models for efficient frontal face synthesis. Our method extracts identity-aware segmentation embeddings and manipulates the corresponding segmentation masks to generate multiple realistic frontal views, selecting the optimal output based on identity loss. The proposed approach demonstrates strong robustness even in extreme cases with severely limited facial input, while requiring only minimal fine-tuning, thereby offering both effectiveness and computational efficiency.

We introduce InsideOut, an extension of 3D Gaussian splatting (3DGS) that bridges the gap between high-fidelity RGB surface details and subsurface X-ray structures. The fusion of RGB and X-ray imaging is invaluable in fields such as medical diagnostics, cultural heritage restoration, and manufacturing. We collect new paired RGB and X-ray data, perform hierarchical fitting to align RGB and X-ray radiative Gaussian splats, and propose an X-ray reference loss to ensure consistent internal structures. InsideOut effectively addresses the challenges posed by disparate data representations between the two modalities and limited paired datasets. This approach significantly extends the applicability of 3DGS, enhancing visualization, simulation, and non-destructive testing capabilities across various domains.

While controllers can support many generic 3D interactions in virtual reality (VR), it alone may fall short of eliciting the core experience for certain actions. One such task is the “object throwing’’, ubiquitous in many sports contents, involving intricately timed actions of aiming, arm swinging, and object releasing. With the increasing availability of eye tracking, we propose to combine gaze-based targeting with the controller swing gesture to model the object throw. The target is aimed/locked by gaze, and the throw is enacted by the controller swing/button press with the object let-gone by the button release. We compare and evaluate the proposed interface against the conventional controller-only based interaction through two typical baseball tasks – Pitcher and Outfielder. The findings indicated that the task accuracy was similar, but the gaze-based targeting allowed for faster completion. More importantly, the gaze-based method showed significantly higher usability and richer VR experience.

Data centers frequently face significant memory under-utilization due to factors such as infrastructure overprovisioning, inefficient workload scheduling, and limited server configurations. This paper introduces Agile-DRAM, a novel DRAM architecture that addresses this issue by flexibly converting the under-utilized memory capacity into enhanced latency performance and reduced power consumption. Through minor modifications to the conventional DRAM architecture, Agile-DRAM supports multiple operational modes: low-latency, lowpower, and the default max-capacity mode. Notably, Agile-DRAM facilitates agile transitions between these modes in response to workload fluctuations in data centers at runtime. Evaluation results demonstrate that the low-latency mode can boost singlecore execution speed by up to 25.8% and diminish energy usage by up to 22.4%. Similarly, the low-power mode can reduce DRAM standby and self-refresh power by 31.6% and 85.7%, respectively.

Locomotion is an important task in many virtual reality (VR) applications. Locomotion requires two different directional settings for: (1) translation to a target, and (2) body rotation. Conventional locomotion methods mostly make use of the hand-held controller for such directional control. The controller-based directional setting may not be natural, as in real life, humans set directions with the gaze, often together with the head/body rotation. Note that the view direction is independently and naturally controlled by the head (or gaze) direction. In this paper, we propose to use the peripheral gaze for body rotation, seamlessly together with the foveal gaze for the view control, so that the two tasks practically do not interfere with one another. In the comparative study, the peripheral gaze showed similar task performance to the controller-based rotation but with reduced VR sickness and improved usability.

DRAM vendors utilize On-Die Error Correction Codes (OD-ECC) to correct random bit errors internally. Meanwhile, system companies utilize Rank-Level ECC (RL-ECC) to protect data against chip errors. Separate protection increases the redundancy ratio to 32.8% in DDR5 and incurs significant performance penalties. This paper proposes a novel RL-ECC, Unity ECC, that can correct both singlechip and double-bit error patterns. Unity ECC corrects doublebit errors using unused syndromes of single-chip correction. Our evaluation shows that Unity ECC without OD-ECC can provide the same reliability level as Chipkill RL-ECC with OD-ECC. Moreover, it can significantly improve system performance and reduce DRAM energy and area by eliminating OD-ECC.