Exploring the Versal AI Engine for 3D Gaussian Splatting

2025-02-17

AI Summary

This paper addresses the challenge of efficiently mapping 3D Gaussian Splatting—a key neural rendering technique—onto AMD’s Versal AI Engine, a spatial architecture with heterogeneous VLIW-SIMD cores and a 2D dataflow network. Method: We present the first systematic analysis of the architecture’s suitability for Gaussian rasterization workloads, and propose a fine-grained task partitioning and cross-core pipelined scheduling strategy that jointly optimizes spherical harmonic color evaluation and covariance matrix computation. Our approach tightly integrates instruction-level parallelism in VLIW-SIMD units with spatial parallelism in the 2D dataflow network, leveraging explicit dataflow configuration and multi-core pipeline coordination. Contribution/Results: The design achieves a 226× throughput improvement over naive mapping, significantly enhancing hardware utilization. It establishes the first reusable, high-performance implementation paradigm for real-time neural rendering on AI Engines.

Technology Category

• Machine Learning: Hardware-aware ML
• Computer Vision: 3D Computer Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Dataflow-oriented spatial architectures are the emerging paradigm for higher computation performance and efficiency. AMD Versal AI Engine is a commercial spatial architecture consisting of tiles of VLIW processors supporting SIMD operations arranged in a two-dimensional mesh. The architecture requires the explicit design of task assignments and dataflow configurations for each tile to maximize performance, demanding advanced techniques and meticulous design. However, a few works revealed the performance characteristics of the Versal AI Engine through practical workloads. In this work, we provide the comprehensive performance evaluation of the Versal AI Engine using Gaussian feature computation in 3D Gaussian splatting as a practical workload, and we then propose a novel dedicated algorithm to fully exploit the hardware architecture. The computations of 3D Gaussian splatting include matrix multiplications and color computations utilizing high-dimensional spherical harmonic coefficients. These tasks are processed efficiently by leveraging the SIMD capabilities and their instruction-level parallelism. Additionally, pipelined processing is achieved by assigning different tasks to individual cores, thereby fully exploiting the spatial parallelism of AI Engines. The proposed method demonstrated a 226-fold throughput increase in simulation-based evaluation, outperforming a naive approach. These findings provide valuable insights for application development that effectively harnesses the spatial and architectural advantages of AI Engines.

Problem

• Evaluate AMD Versal AI Engine performance
• Optimize 3D Gaussian splatting computations
• Enhance spatial parallelism in AI Engines

Innovation

• Utilizes AMD Versal AI Engine
• Optimizes SIMD and parallelism
• Achieves pipelined task processing