Rendering 3D Gaussians on a Graph Processor

• URL: http://arxiv.org/abs/2607.15951v1
• Date: Fri, 17 Jul 2026 13:36:30 GMT
• Title: Rendering 3D Gaussians on a Graph Processor
• Abstract summary: First implementation of a 3D Gaussian on an Intelligence Processing Unit (IPU)<n>1,472 independent tiles with only on-chip; constraints approximate properties of efficient sensor-processor architectures.<n>We evaluate how these constraints affect performance and render quality.
• Score: 26.271402958805766
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present the first implementation of a 3D Gaussian renderer on an Intelligence Processing Unit (IPU), comprising 1,472 independent tiles with only on-chip SRAM; constraints that approximate properties of efficient sensor-processor architectures. Our input scenes are 3D Gaussian maps from real-world sequences. Each tile 'owns' a screen-space region of the framebuffer; Gaussian primitives are routed to destination tiles via Manhattan-distance hops on a north-east-west-south (NEWS) grid, then distributed to overlapping neighbours in an expanding tree pattern. Computation follows the IPU's Bulk Synchronous Parallel (BSP) model, with inter-tile communication defined at compile time. We show this hardware allows us to exploit spatial and temporal locality by enabling local data transfer between cores. We evaluate the bottlenecks in this SRAM-only implementation: inter-tile bandwidth, per-tile SRAM capacity, and workload imbalance from non-uniform Gaussian density. We analyse how these constraints affect performance and render quality. This exploration raises broader questions for conventional GPUs and 3D representations, suggesting that direct inter-SM (streaming multiprocessor) communication might offer ways to reduce DRAM access in GPU kernels. We discuss these implications for the future of on-sensor and DRAM-free architectures. Project page: https://nmjfry.github.io/ipu-3dgs/

Related papers

• Local-GS: Accelerating 3D Gaussian Splatting via Tile-Local Warp Coherence [31.001161060131622]
  3D Gaussian Splatting (3DGS) has significantly advanced real-time novel view computation by representing scenes as dense collections of anisotropic 3D Gaussian primitives.<n>We present Local-GS, a warp-coherent rendering paradigm that organizes Gaussian primitives with respect to SIMT (Single Instruction, Multiple Threads) execution boundaries rather than scene geometry.<n>As a plug-and-play optimization, it provides additional performance gains to all tested baselines, culminating in a $7.76times$ speedup on Deep Blending scenes.
  arXiv  Detail & Related papers  (2026-06-15T11:08:10Z)
• Accelerating GPU Inference of Large Language Models with Moderately Unstructured Sparse Weight Matrices [18.428868416628017]
  This paper proposes an efficient inference method for large language models (LLMs) with moderate sparsity.<n>It achieves up to 1.64x-level speedup over SpInfer and up to 1.41x end-to-end speedups over FlashLLM.
  arXiv  Detail & Related papers  (2026-06-13T13:38:27Z)
• Radiant Foam Rendering on a Graph Processor [25.181643235594734]
  Many-core accelerators replace a single large device memory with hundreds to thousands of lightweight cores, each owning only a small local and exchanging data via explicit on-chip communication.<n>This organization offers high aggregate bandwidth, but it breaks a key assumption behind many techniques: that rays can randomly access a large, unified scene representation.<n>We present a fully in rendering-SRAM, distributed for the Radiant Foam Voronoi-cell volumetric representation on the Graphcore Mk2 IPU.<n>Our system attains throughput of approximately 1 fps at 640x480 with image and depth map quality close to the original GPU-based Radiant Foam implementation
  arXiv  Detail & Related papers  (2026-01-07T20:44:04Z)
• Quantile Rendering: Efficiently Embedding High-dimensional Feature on 3D Gaussian Splatting [52.18697134979677]
  Recent advancements in computer vision have successfully extended Open-vocabulary segmentation (OVS) to the 3D domain by leveraging 3D Gaussian Splatting (3D-GS)<n>Existing methods employ codebooks or feature compression, causing information loss, thereby degrading segmentation quality.<n>We introduce Quantile Rendering (Q-Render), a novel rendering strategy for 3D Gaussians that efficiently handles high-dimensional features while maintaining high fidelity.<n>Our framework outperforms state-of-the-art methods, while enabling real-time rendering with an approximate 43.7x speedup on 512-D feature maps.
  arXiv  Detail & Related papers  (2025-12-24T04:16:18Z)
• LODGE: Level-of-Detail Large-Scale Gaussian Splatting with Efficient Rendering [75.67501939005119]
  We present a novel level-of-detail (LOD) method for 3D Gaussian Splatting on memory-constrained devices.<n>Our approach iteratively selects optimal subsets of Gaussians based on camera distance.<n>Our method achieves state-of-the-art performance on both outdoor (Hierarchical 3DGS) and indoor (Zip-NeRF) datasets.
  arXiv  Detail & Related papers  (2025-05-29T06:50:57Z)
• GaussRender: Learning 3D Occupancy with Gaussian Rendering [86.89653628311565]
  GaussRender is a module that improves 3D occupancy learning by enforcing projective consistency.<n>Our method penalizes 3D configurations that produce inconsistent 2D projections, thereby enforcing a more coherent 3D structure.
  arXiv  Detail & Related papers  (2025-02-07T16:07:51Z)
• Deep Optimizer States: Towards Scalable Training of Transformer Models Using Interleaved Offloading [2.8231000588510757]
  Transformers and large language models(LLMs) have seen rapid adoption in all domains. Training of transformers is very expensive and often hits a memory wall'' We propose a novel technique to split the LLM into subgroups, whose update phase is scheduled on either the CPU or the GPU.
  arXiv  Detail & Related papers  (2024-10-26T00:43:59Z)
• SplatLoc: 3D Gaussian Splatting-based Visual Localization for Augmented Reality [50.179377002092416]
  We propose an efficient visual localization method capable of high-quality rendering with fewer parameters. Our method achieves superior or comparable rendering and localization performance to state-of-the-art implicit-based visual localization approaches.
  arXiv  Detail & Related papers  (2024-09-21T08:46:16Z)
• Practical offloading for fine-tuning LLM on commodity GPU via learned sparse projectors [11.127604539303373]
  Fine-tuning large language models (LLMs) requires significant memory, often exceeding the capacity of a single GPU.<n>A common solution to this memory challenge is offloading compute and data from the GPU to the CPU.<n>We present an offloading framework, LSP-Offload, that enables near-native speed LLM fine-tuning on commodity hardware.
  arXiv  Detail & Related papers  (2024-06-14T16:59:11Z)
• SplaTAM: Splat, Track & Map 3D Gaussians for Dense RGB-D SLAM [48.190398577764284]
  SplaTAM is an approach to enable high-fidelity reconstruction from a single unposed RGB-D camera. It employs a simple online tracking and mapping system tailored to the underlying Gaussian representation. Experiments show that SplaTAM achieves up to 2x superior performance in camera pose estimation, map construction, and novel-view synthesis over existing methods.
  arXiv  Detail & Related papers  (2023-12-04T18:53:24Z)
• INR-Arch: A Dataflow Architecture and Compiler for Arbitrary-Order Gradient Computations in Implicit Neural Representation Processing [66.00729477511219]
  Given a function represented as a computation graph, traditional architectures face challenges in efficiently computing its nth-order gradient. We introduce INR-Arch, a framework that transforms the computation graph of an nth-order gradient into a hardware-optimized dataflow architecture. We present results that demonstrate 1.8-4.8x and 1.5-3.6x speedup compared to CPU and GPU baselines respectively.
  arXiv  Detail & Related papers  (2023-08-11T04:24:39Z)
• Mesh Convolution with Continuous Filters for 3D Surface Parsing [101.25796935464648]
  We propose a series of modular operations for effective geometric feature learning from 3D triangle meshes. Our mesh convolutions exploit spherical harmonics as orthonormal bases to create continuous convolutional filters. We further contribute a novel hierarchical neural network for perceptual parsing of 3D surfaces, named PicassoNet++.
  arXiv  Detail & Related papers  (2021-12-03T09:16:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.