FlashFPS: Efficient Farthest Point Sampling for Large-Scale Point Clouds via Pruning and Caching

• URL: http://arxiv.org/abs/2604.17720v1
• Date: Mon, 20 Apr 2026 02:03:24 GMT
• Title: FlashFPS: Efficient Farthest Point Sampling for Large-Scale Point Clouds via Pruning and Caching
• Abstract summary: Point-based Neural Networks (PNNs) have become a key approach for point cloud processing.<n>Farthest Point Sampling (FPS) often introduces significant latency.<n>FPS remains a major bottleneck due to exhaustive computations across multiple network layers in PNNs.
• Score: 10.428491193699374
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Point-based Neural Networks (PNNs) have become a key approach for point cloud processing. However, a core operation in these models, Farthest Point Sampling (FPS), often introduces significant inference latency, especially for large-scale processing. Despite existing CUDA- and hardware-level optimizations, FPS remains a major bottleneck due to exhaustive computations across multiple network layers in PNNs, which hinders scalability. Through systematic analysis, we identify three substantial redundancies in FPS, including unnecessary full-cloud computations, redundant late-stage iterations, and predictable inter-layer outputs that make later FPS computations avoidable. To address these, we propose \textbf{\textit{FlashFPS}}, a hardware-agnostic, plug-and-play framework for FPS acceleration, composed of \textit{FPS-Prune} and \textit{FPS-Cache}. \textit{FPS-Prune} introduces candidate pruning and iteration pruning to reduce redundant computations in FPS while preserving sampling quality, and \textit{FPS-Cache} eliminates layer-wise redundancy via cache-and-reuse. Integrated into existing CUDA libraries and state-of-the-art PNN accelerators, \textit{FlashFPS} achieves 5.16$\times$ speedup over the standard CUDA baseline on GPU and 2.69$\times$ on PNN accelerators, with negligible accuracy loss, enabling efficient and scalable PNN inference. Codes are released at https://github.com/Yuzhe-Fu/FlashFPS.

Related papers

• FAST: A Holistic Framework for Optimizing Memory-I/O, Computation, and Sampling in Temporal GNN Training [13.505374557741169]
  Temporal Graph Neural Networks (TGNNs) are widely used for learning from dynamic graphs in applications such as recommendation, social network analysis, and traffic forecasting.<n>We present FAST, a holistic framework that accelerates end-to-end TGNN training by jointly optimizing sampling, memory I/O, and computation.<n>FAST achieves an average of 2.1x (up to 4.7x) speedup over state-of-the-art systems without sacrificing model accuracy.
  arXiv  Detail & Related papers  (2026-07-06T13:54:01Z)
• RadiusFPS: Efficient Farthest Point Sampling on CPUs and GPUs via Spherical Voxel Pruning [7.454629583409613]
  Point clouds are a primary sensory representation for robotic perception.<n>FPS is the most well-known downsampling operator.<n>FPS scales poorly with the million-point-per-second rates of modern 3D sensors.
  arXiv  Detail & Related papers  (2026-06-04T14:57:05Z)
• LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation [84.86814271505109]
  LongLive-2.0 is an NVFP4-based parallel infrastructure throughout the full training and inference workflow of long video generation.<n>For training, we introduce sequence-parallel autoregressive (AR) training, instantiated as Balanced SP.<n>Experiments show up to 2.15x speedup in training, and 1.84x in inference.
  arXiv  Detail & Related papers  (2026-05-18T17:57:03Z)
• Spava: Accelerating Long-Video Understanding via Sequence-Parallelism-aware Approximate Attention [63.69228529380251]
  Spava is a sequence-parallel framework with optimized attention for long-video inference.<n>Spava delivers speedups of 12.72x, 1.70x, and 1.18x over FlashAttn, ZigZagRing, and APB, without notable performance loss.
  arXiv  Detail & Related papers  (2026-01-29T09:23:13Z)
• HLS4PC: A Parametrizable Framework For Accelerating Point-Based 3D Point Cloud Models on FPGA [2.762332196350206]
  3D point cloud models execute computation and memory intensive mapping functions alongside NN layers for classification/segmentation.<n>PointMLP-Lite is a 4x less complex variant with only 2% accuracy drop on ModelNet40.<n>Our implementation achieves 2.3x and 22x higher throughput compared to the GPU and CPU implementations, respectively.
  arXiv  Detail & Related papers  (2025-12-11T17:09:12Z)
• FractalCloud: A Fractal-Inspired Architecture for Efficient Large-Scale Point Cloud Processing [13.217596969807062]
  Three-dimensional (3D) point clouds are increasingly used in applications such as autonomous driving, robotics, and virtual reality (VR)<n>Point-based neural networks (PNNs) have demonstrated strong performance in point cloud analysis, originally targeting small-scale inputs.<n>FractalCloud is a fractal-inspired hardware architecture for efficient large-scale 3D point cloud processing.
  arXiv  Detail & Related papers  (2025-11-10T22:19:37Z)
• ReTiDe: Real-Time Denoising for Energy-Efficient Motion Picture Processing with FPGAs [0.39146761527401425]
  This paper presents Real-Time Denoise (ReTiDe), a hardware-accelerated denoising system that serves inference on datacentre FPGAs.<n>A compact convolutional model is quantised (post-training quantisation plus quantisation-aware fine-tuning) to INT8 and compiled for AMD Deep Learning Processor Unit (DPU)-based FPGAs.<n>A client-server integration offloads from the CPU/ GPU to a networked FPGA service, while remaining callable from existing host, e.g., NUKE, without disrupting artist tooling.
  arXiv  Detail & Related papers  (2025-10-04T13:43:43Z)
• H2PIPE: High throughput CNN Inference on FPGAs with High-Bandwidth Memory [1.0056445773367833]
  Convolutional Neural Networks (CNNs) combine large amounts of parallelizable computation with frequent memory access. This work augments a state-of-the-art dataflow accelerator to leverage both High-Bandwidth Memory (HBM) and on-chip storage. Compared to the best prior work we obtain speed-ups of at least 19.4x, 5.1x and 10.5x on ResNet-18, ResNet-50 and VGG-16 respectively.
  arXiv  Detail & Related papers  (2024-08-17T14:25:32Z)
• Adaptable Butterfly Accelerator for Attention-based NNs via Hardware and Algorithm Co-design [66.39546326221176]
  Attention-based neural networks have become pervasive in many AI tasks. The use of the attention mechanism and feed-forward network (FFN) demands excessive computational and memory resources. This paper proposes a hardware-friendly variant that adopts a unified butterfly sparsity pattern to approximate both the attention mechanism and the FFNs.
  arXiv  Detail & Related papers  (2022-09-20T09:28:26Z)
• Sparse Periodic Systolic Dataflow for Lowering Latency and Power Dissipation of Convolutional Neural Network Accelerators [3.043665249713003]
  This paper introduces the sparse periodic systolic (SPS) dataflow, which advances the state-of-the-art hardware accelerator for supporting lightweight neural networks. By exploiting the regularity of PPS, our sparsity-aware compiler optimally reorders the weights and uses a simple indexing unit in hardware to create matches between the weights and activations.
  arXiv  Detail & Related papers  (2022-06-30T19:16:46Z)
• Instant Neural Graphics Primitives with a Multiresolution Hash Encoding [67.33850633281803]
  We present a versatile new input encoding that permits the use of a smaller network without sacrificing quality. A small neural network is augmented by a multiresolution hash table of trainable feature vectors whose values are optimized through a gradient descent. We achieve a combined speed of several orders of magnitude, enabling training of high-quality neural graphics primitives in a matter of seconds.
  arXiv  Detail & Related papers  (2022-01-16T07:22:47Z)
• An Adaptive Device-Edge Co-Inference Framework Based on Soft Actor-Critic [72.35307086274912]
  High-dimension parameter model and large-scale mathematical calculation restrict execution efficiency, especially for Internet of Things (IoT) devices. We propose a new Deep Reinforcement Learning (DRL)-Soft Actor Critic for discrete (SAC-d), which generates the emphexit point, emphexit point, and emphcompressing bits by soft policy iterations. Based on the latency and accuracy aware reward design, such an computation can well adapt to the complex environment like dynamic wireless channel and arbitrary processing, and is capable of supporting the 5G URL
  arXiv  Detail & Related papers  (2022-01-09T09:31:50Z)
• FastFlowNet: A Lightweight Network for Fast Optical Flow Estimation [81.76975488010213]
  Dense optical flow estimation plays a key role in many robotic vision tasks. Current networks often occupy large number of parameters and require heavy computation costs. Our proposed FastFlowNet works in the well-known coarse-to-fine manner with following innovations.
  arXiv  Detail & Related papers  (2021-03-08T03:09:37Z)
• EdgeBERT: Sentence-Level Energy Optimizations for Latency-Aware Multi-Task NLP Inference [82.1584439276834]
  Transformer-based language models such as BERT provide significant accuracy improvement for a multitude of natural language processing (NLP) tasks. We present EdgeBERT, an in-depth algorithm- hardware co-design for latency-aware energy optimization for multi-task NLP.
  arXiv  Detail & Related papers  (2020-11-28T19:21:47Z)

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.