3D Gaussian Splatting can exploit frustum culling and level-of-detail strategies to accelerate rendering of scenes containing a large number of primitives. However, the semi-transparent nature of Gaussians prevents the application of another highly effective technique: occlusion culling. We address this limitation by proposing a novel method to learn the viewpoint-dependent visibility function of all Gaussians in a trained model using a small, shared MLP across instances of an asset in a scene. By querying it for Gaussians within the viewing frustum prior to rasterization, our method can discard occluded primitives during rendering. Leveraging Tensor Cores for efficient computation, we integrate these neural queries directly into a novel instanced software rasterizer. Our approach outperforms the current state of the art for composed scenes in terms of VRAM usage and image quality, utilizing a combination of our instanced rasterizer and occlusion culling MLP, and exhibits complementary properties to existing LoD techniques.
Robust Visual SLAM (vSLAM) is essential for autonomous systems operating in real-world environments, where challenges such as dynamic objects, low texture, and critically, varying illumination conditions often degrade performance. Existing feature-based SLAM systems rely on fixed front-end parameters, making them vulnerable to sudden lighting changes and unstable feature tracking. To address these challenges, we propose ``IRAF-SLAM'', an Illumination-Robust and Adaptive Feature-Culling front-end designed to enhance vSLAM resilience in complex and challenging environments. Our approach introduces: (1) an image enhancement scheme to preprocess and adjust image quality under varying lighting conditions; (2) an adaptive feature extraction mechanism that dynamically adjusts detection sensitivity based on image entropy, pixel intensity, and gradient analysis; and (3) a feature culling strategy that filters out unreliable feature points using density distribution analysis and a lighting impact factor. Comprehensive evaluations on the TUM-VI and European Robotics Challenge (EuRoC) datasets demonstrate that IRAF-SLAM significantly reduces tracking failures and achieves superior trajectory a
While Artificial Intelligence (AI) is not a new field, recent developments, especially with the release of generative tools like ChatGPT, have brought it to the forefront of the minds of industry workers and academic folk alike. There is currently much talk about AI and its ability to reshape many everyday processes as we know them through automation. It also allows users to expand their ideas by suggesting things they may not have thought of on their own and provides easier access to information. However, not all of the changes this technology will bring or has brought so far are positive; this is why it is extremely important for all modern people to recognize and understand the risks before using these tools and allowing them to cause harm. This work takes a position on better understanding many equity concerns and the spread of misinformation that result from new AI, in this case, specifically ChatGPT and deepfakes, and encouraging collaboration with law enforcement, developers, and users to reduce harm. Considering many academic sources, it warns against these issues, analyzing their cause and impact in fields including healthcare, education, science, academia, retail, and fin
The emerging threat of a human pandemic caused by the H5N1 avian influenza virus strain magnifies the need for controlling the incidence of H5N1 infection in domestic bird populations. Culling is one of the most widely used control measures and has proved effective for isolated outbreaks. However, the socio-economic impacts of mass culling, in the face of a disease which has become endemic in many regions of the world, can affect the implementation and success of culling as a control measure. We use mathematical modeling to understand the dynamics of avian influenza under different culling approaches. We incorporate culling into an SI model by considering the per capita culling rates to be general functions of the number of infected birds. Complex dynamics of the system, such as backward bifurcation and forward hysteresis, along with bi-stability, are detected and analyzed for two distinct culling scenarios. In these cases, employing other control measures temporarily can drastically change the dynamics of the solutions to a more favorable outcome for disease control.
Ray tracing is an essential operation for realistic image synthesis. The acceleration of ray tracing has been studied for a long period of time because algorithms such as light transport simulations require a large amount of ray tracing. One of the major approaches to accelerate the intersections is to use bounding volumes for early pruning for primitives in the volume. The axis-aligned bounding box is a popular bounding volume for ray tracing because of its simplicity and efficiency. However, the conservative bounding volume may produce extra empty space in addition to its content. Especially, primitives that are thin and diagonal to the axis give false-positive hits on the box volume due to the extra space. Although more complex bounding volumes such as oriented bounding boxes may reduce more false-positive hits, they are computationally expensive. In this paper, we propose a novel culling approach to reduce false-positive hits for the bounding box by embedding a binary voxel data structure to the volume. As a ray is represented as a conservative voxel volume as well in our approach, the ray--voxel intersection is cheaply done by bitwise AND operations. Our method is applicable t
Real-time CNN-based object detection models for applications like surveillance can achieve high accuracy but are computationally expensive. Recent works have shown 10 to 100x reduction in computation cost for inference by using domain-specific networks. However, prior works have focused on inference only. If the domain model requires frequent retraining, training costs can pose a significant bottleneck. To address this, we propose Dataset Culling: a pipeline to reduce the size of the dataset for training, based on the prediction difficulty. Images that are easy to classify are filtered out since they contribute little to improving the accuracy. The difficulty is measured using our proposed confidence loss metric with little computational overhead. Dataset Culling is extended to optimize the image resolution to further improve training and inference costs. We develop fixed-angle, long-duration video datasets across several domains, and we show that the dataset size can be culled by a factor of 300x to reduce the total training time by 47x with no accuracy loss or even with slight improvement. Codes are available: https://github.com/kentaroy47/DatasetCulling
Highly Pathogenic Avian Influenza A H5N6 is a mutated virus of Influenza A H5N1 and a new emerging infection that recently caused an outbreak in the Philippines. The 2017 H5N6 outbreak resulted in a depopulation of 667,184 domestic birds. In this study, we incorporate half-saturated incidence in our mathematical models and investigate three intervention strategies against H5N6: isolation with treatment, vaccination and modified culling. We determine the direction of the bifurcation when $\mathcal{R}_0 = 1$ and show that all the models exhibit forward bifurcation. We administer optimal control and perform numerical simulations to compare the consequences and implementation cost of utilizing different intervention strategies in the poultry population. Despite the challenges of applying each control strategy, we show that culling both infected and susceptible birds is a better control strategy in prohibiting an outbreak and avoiding further recurrence of the infection from the population compared to confinement and vaccination.
For a continuous-time Bienaymé-Galton-Watson process, $X$, with immigration and culling, $0$ as an absorbing state, call $X^q$ the process that results from killing $X$ at rate $q\in (0,\infty)$, followed by stopping it on extinction or explosion. Then an explicit identification of the relevant harmonic functions of $X^q$ allows to determine the Laplace transforms (at argument $q$) of the first passage times downwards and of the explosion time for $X$. Strictly speaking, this is accomplished only when the killing rate $q$ is sufficiently large (but always when the branching mechanism is not supercritical or if there is no culling). In particular, taking the limit $q\downarrow 0$ (whenever possible) yields the passage downwards and explosion probabilities for $X$. A number of other consequences of these results are presented.
Randomized parallel algorithms for many fundamental problems achieve optimal linear work in expectation, but upgrading this guarantee to hold with high probability (whp) remains a recurring theoretical challenge. In this paper, we address this gap for several core parallel primitives. First, we present the first parallel semisort algorithm achieving $O(n)$ work and $O(\text{polylog } n)$ depth whp, improving upon the $O(n)$ expected work bound of Gu et al. [SPAA 2015]. Our analysis introduces new concentration arguments based on simple tabulation hashing and tail bounds for weighted sums of geometric random variables. As a corollary, we obtain an integer sorting algorithm for keys in $[n]$ matching the same bounds. Second, we introduce a framework for boosting randomized parallel graph algorithms from expected to high probability linear work. The framework applies to \emph{locally extendable} problems -- those admitting a deterministic procedure that extends a solution across a graph cut in work proportional to the cut size. We combine this with a \emph{culled balanced partition} scheme: an iterative culling phase removes a polylogarithmic number of high-degree vertices, after whic
Recent advancements in Gaussian Splatting (3DGS) have introduced various modifications to the original kernel, resulting in significant performance improvements. However, many of these kernel changes are incompatible with existing datasets optimized for the original Gaussian kernel, presenting a challenge for widespread adoption. In this work, we address this challenge by proposing an alternative kernel that maintains compatibility with existing datasets while improving computational efficiency. Specifically, we replace the original exponential kernel with a polynomial approximation combined with a ReLU function. This modification allows for more aggressive culling of Gaussians, leading to enhanced performance across different 3DGS implementations. Our results show a notable performance improvement of 4 to 15% with negligible impact on image quality. We also provide a detailed mathematical analysis of the new kernel and discuss its potential benefits for 3DGS implementations on NPU hardware.
We present WebSplatter, an end-to-end GPU rendering pipeline for the heterogeneous web ecosystem. Unlike naive ports, WebSplatter introduces a wait-free hierarchical radix sort that circumvents the lack of global atomics in WebGPU, ensuring deterministic execution across diverse hardware. Furthermore, we propose an opacity-aware geometry culling stage that dynamically prunes splats before rasterization, significantly reducing overdraw and peak memory footprint. Evaluation demonstrates that WebSplatter consistently achieves 1.2$\times$ to 4.5$\times$ speedups over state-of-the-art web viewers.
We introduce a probabilistic splat-based radiance field framework that retains the fast rasterization and test-time efficiency of 3D Gaussian Splatting (3DGS) while replacing heuristic primitive manipulation with gradient-based optimization of a volumetric probability density. Rather than relocating, splitting, or culling Gaussians via hand-tuned densification (e.g., ADC), we treat primitive locations as samples drawn from a persistent, learnable density. We instantiate this density using a novel, memory-efficient multi-scale hierarchical grid that enables end-to-end gradient-based optimization. To stabilize the optimization, we derive an unbiased gradient estimator with control variates that markedly reduces variance. By allowing probability mass to flow to where the loss demands, our framework eliminates brittle priors and naturally explores the volume, achieving state-of-the-art reconstruction quality on mip-NeRF 360 while preserving 3DGS-level rendering speed.
Real-time Light Detection And Ranging (LiDAR) simulation must find, per emitted ray, the closest intersecting triangle even in dynamic scenes containing large numbers of moving and deformable objects. Dominant acceleration-structure approaches require rebuilding each frame for dynamic geometry -- a cost that compounds directly with scene dynamics and cannot be amortized regardless of how little actually changed. This paper presents the Gajmer Ray-Casting Algorithm (GRCA), which inverts the question: instead of asking what does each ray hit? it asks which rays can each triangle possibly hit? GRCA geometrically models spinning LiDAR emitters as rotation-traced cones or planes and uses each triangle's emitter-centric apparent area to cull, per triangle, which channels and the rays within those channels can possibly reach it -- without any acceleration structure. GRCA is compute-based and vendor-agnostic by design, targeting highly dynamic, high-resolution simultaneous multi-sensor simulation. At its core, GRCA is a general-purpose ray-casting algorithm: the emitter-centric inversion applies to any setting where rays originate from a known position, not only LiDAR. Benchmarks evaluate
Visualizing large 3D scientific datasets requires balancing performance and fidelity, but traditional tools often demand excessive technical expertise. We introduce UnrealVis, an Unreal Engine optimization laboratory for configuring and evaluating rendering techniques during interactive exploration. Following a review of 55 papers, we established a taxonomy of 22 optimization techniques across six families, implementing them through engine subsystems such as Nanite, Level of Detail(LOD) schemes, and culling. The system features an intuitive workflow with live telemetry and A/B comparisons for local and global performance analysis. Validated through case studies of ribosomal structures and volumetric flow fields, along with an expert evaluation, UnrealVis facilitates the selection of optimization combinations that meet performance goals while preserving structural fidelity. UnrealVis is available at https://github.com/XAIber-lab/UnrealVis
The advent of 3D Gaussian Splatting has revolutionized graphics rendering by delivering high visual quality and fast rendering speeds. However, training large-scale scenes at high quality remains challenging due to the substantial memory demands required to store parameters, gradients, and optimizer states, which can quickly overwhelm GPU memory. To address these limitations, we propose GS-Scale, a fast and memory-efficient training system for 3D Gaussian Splatting. GS-Scale stores all Gaussians in host memory, transferring only a subset to the GPU on demand for each forward and backward pass. While this dramatically reduces GPU memory usage, it requires frustum culling and optimizer updates to be executed on the CPU, introducing slowdowns due to CPU's limited compute and memory bandwidth. To mitigate this, GS-Scale employs three system-level optimizations: (1) selective offloading of geometric parameters for fast frustum culling, (2) parameter forwarding to pipeline CPU optimizer updates with GPU computation, and (3) deferred optimizer update to minimize unnecessary memory accesses for Gaussians with zero gradients. Our extensive evaluations on large-scale datasets demonstrate tha
Dynamic 3D Gaussian splatting (3DGS) extends static 3DGS to render dynamic scenes, enabling AR/VR applications with moving objects. However, implementing dynamic 3DGS on edge devices faces challenges: (1) Loading all Gaussian parameters from DRAM for frustum culling incurs high energy costs. (2) Increased parameters for dynamic scenes elevate sorting latency and energy consumption. (3) Limited on-chip buffer capacity with higher parameters reduces buffer reuse, causing frequent DRAM access. (4) Dynamic 3DGS operations are not readily compatible with digital compute-in-memory (DCIM). These challenges hinder real-time performance and power efficiency on edge devices, leading to reduced battery life or requiring bulky batteries. To tackle these challenges, we propose algorithm-hardware co-design techniques. At the algorithmic level, we introduce three optimizations: (1) DRAM-access reduction frustum culling to lower DRAM access overhead, (2) Adaptive tile grouping to enhance on-chip buffer reuse, and (3) Adaptive interval initialization Bucket-Bitonic sort to reduce sorting latency. At the hardware level, we present a DCIM-friendly computation flow that is evaluated using the measured
For two resource-sharing species we explore the interplay of harvesting and dispersal strategies, as well as their influence on competition outcomes. Although the extinction of either species can be achieved by excessive culling, choosing a harvesting strategy such that the biodiversity of the populations is preserved is much more complicated. We propose a type of heterogeneous harvesting policy, dependent on dispersal strategy, where the two managed populations become an ideal free pair, and show that this strategy guarantees the coexistence of the species. We also show that if the harvesting of one of the populations is perturbed in some way, then it is possible for the coexistence to be preserved. Further, we show that if the dispersal of two species formed an ideal free pair, then a slight change in the dispersal strategy for one of them does not affect their ability to coexist. Finally, in the model, directed movement is represented by the term $Δ(u/P)$, where $P$ is the dispersal strategy and target distribution. We justify that once an invading species, which has an advantage in carrying capacity, chooses a dispersal strategy that mimics the resident species distribution, th
3D Gaussian Splatting (3DGS) is a recent explicit 3D representation that has achieved high-quality reconstruction and real-time rendering of complex scenes. However, the rasterization pipeline still suffers from unnecessary overhead resulting from avoidable serial Gaussian culling, and uneven load due to the distinct number of Gaussian to be rendered across pixels, which hinders wider promotion and application of 3DGS. In order to accelerate Gaussian splatting, we propose AdR-Gaussian, which moves part of serial culling in Render stage into the earlier Preprocess stage to enable parallel culling, employing adaptive radius to narrow the rendering pixel range for each Gaussian, and introduces a load balancing method to minimize thread waiting time during the pixel-parallel rendering. Our contributions are threefold, achieving a rendering speed of 310% while maintaining equivalent or even better quality than the state-of-the-art. Firstly, we propose to early cull Gaussian-Tile pairs of low splatting opacity based on an adaptive radius in the Gaussian-parallel Preprocess stage, which reduces the number of affected tile through the Gaussian bounding circle, thus reducing unnecessary ove
In this work we analyze and address a fundamental restriction that blocks the reliable application of codimensional yarn-level and shell models with thickness, to simulate real-world woven and knit fabrics. As discretizations refine toward practical and accurate physical modeling, such models can generate non-physical contact forces with stencil-neighboring elements in the simulation mesh, leading to severe locking artifacts. While not well-documented in the literature, this restriction has so far been addressed with two alternatives with undesirable tradeoffs. One option is to restrict the mesh to coarse resolutions, however, this eliminates the possibility of accurate (and consistent) resolution simulations across real-world material variations. A second alternative instead seeks to cull contact pairs that can create such locking forces in the first place. This relaxes resolution restrictions but compromise robustness. Culling can and will generate unacceptable and unpredictable geometric intersections and tunneling that destroys weaving and knitting structures and cause unrecoverable pull-throughs. We address these challenges to simulating real-world materials with a new and pra
We present an accurate and GPU-accelerated Stereo Visual SLAM design called Jetson-SLAM. It exhibits frame-processing rates above 60FPS on NVIDIA's low-powered 10W Jetson-NX embedded computer and above 200FPS on desktop-grade 200W GPUs, even in stereo configuration and in the multiscale setting. Our contributions are threefold: (i) a Bounded Rectification technique to prevent tagging many non-corner points as a corner in FAST detection, improving SLAM accuracy. (ii) A novel Pyramidal Culling and Aggregation (PyCA) technique that yields robust features while suppressing redundant ones at high speeds by harnessing a GPU device. PyCA uses our new Multi-Location Per Thread culling strategy (MLPT) and Thread-Efficient Warp-Allocation (TEWA) scheme for GPU to enable Jetson-SLAM achieving high accuracy and speed on embedded devices. (iii) Jetson-SLAM library achieves resource efficiency by having a data-sharing mechanism. Our experiments on three challenging datasets: KITTI, EuRoC, and KAIST-VIO, and two highly accurate SLAM backends: Full-BA and ICE-BA show that Jetson-SLAM is the fastest available accurate and GPU-accelerated SLAM system (Fig. 1).