An official website of the United States government
Point cloud is an important type of geometric data structure for many embedded applications such as autonomous driving and augmented reality. Current Point Cloud Networks (PCNs) have proven to achieve great success in using inference to perform point cloud analysis, including object part segmentation, shape classification, and so on. However, point cloud applications on the computing edge require more than just the inference step. They require an end-to-end (E2E) processing of the point cloud workloads: pre-processing of raw data, input preparation, and inference to perform point cloud analysis. Current PCN approaches to support end-to-end processing of point cloud workload cannot meet the real-time latency requirement on the edge, i.e., the ability of the AI service to keep up with the speed of raw data generation by 3D sensors. Latency for end-to-end processing of the point cloud workloads stems from two reasons: memory-intensive down-sampling in the pre-processing phase and the data structuring step for input preparation in the inference phase. In this paper, we present HgPCN, an end-to-end heterogeneous architecture for real-time embedded point cloud applications. In HgPCN, we introduce two novel methodologies based on spatial indexing to address the two identified bottlenecks. In the Pre-processing Engine of HgPCN, an Octree-Indexed-Sampling method is used to optimize the memory-intensive down-sampling bottleneck of the pre-processing phase. In the Inference Engine, HgPCN extends a commercial DLA with a customized Data Structuring Unit which is based on a Voxel-Expanded Gathering method to fundamentally reduce the workload of the data structuring step in the inference phase. The initial prototype of HgPCN has been implemented on an Intel PAC (Xeon+FPGA) platform. Four commonly available point cloud datasets were used for comparison, running on three baseline devices: Intel Xeon W-2255, Nvidia Xavier NX Jetson GPU, and Nvidia 4060ti GPU. These point cloud datasets were also run on two existing PCN accelerators for comparison: PointACC and Mesorasi. Our results show that for the inference phase, depending on the dataset size, HgPCN achieves speedup from 1.3× to 10.2× vs. PointACC, 2.2× to 16.5× vs. Mesorasi, and 6.4× to 21× vs. Jetson NX GPU. Along with optimization of the memory-intensive down-sampling bottleneck in pre-processing phase, the overall latency shows that HgPCN can reach the real-time requirement by providing end-to-end service with keeping up with the raw data generation rate.
more »
« less
TorchSparse++: Efficient Point Cloud Engine
Point cloud computation has become an increasingly more important workload thanks to its applications in autonomous driving. Unlike dense 2D computation, point cloud convolution has sparse and irregular computation patterns and thus requires dedicated inference system support with specialized high-performance kernels. While existing point cloud deep learning libraries have developed different dataflows for convolution on point clouds, they assume a single dataflow throughout the execution of the entire model. In this work, we systematically analyze and improve existing dataflows. Our resulting system, TorchSparse++, achieves 2.9x, 3.3x, 2.2x and 1.8x measured end-to-end speedup on an NVIDIA A100 GPU over the state-of-the-art MinkowskiEngine, SpConv 1.2, TorchSparse and SpConv v2 in inference respectively. Furthermore, TorchSparse++ is the only system to date that supports all necessary primitives for 3D segmentation, detection, and reconstruction workloads in autonomous driving. Code is publicly released at https://github.com/mit-han-lab/torchsparse.
more »
« less
- Award ID(s):
- 2117997
- PAR ID:
- 10477692
- Publisher / Repository:
- https://openaccess.thecvf.com/content/CVPR2023W/WAD/html/Tang_TorchSparse_Efficient_Point_Cloud_Engine_CVPRW_2023_paper.html
- Date Published:
- Journal Name:
- IEEE Xplore digital library
- ISSN:
- 2473-2001
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Recently, point cloud (PC) has gained popularity in modeling various 3D objects (including both synthetic and real-life) and has been extensively utilized in a wide range of applications such as AR/VR, 3D reconstruction, and autonomous driving. For such applications, it is critical to analyze/understand the surrounding scenes properly. To achieve this, deep learning based methods (e.g., convolutional neural networks (CNNs)) have been widely employed for higher accuracy. Unlike the deep learning on conventional 2D images/videos, where the feature computation (matrix multiplication) is the major bottleneck, in point cloud-based CNNs, the sample and neighbor search stages are the primary bottlenecks, and collectively contribute to 54% (up to 80%) of the overall execution latency on a typical edge device. While prior efforts have attempted to solve this issue by designing custom ASICs or pipelining the neighbor search with other stages, to our knowledge, none of them has tried to “structurize” the unstructured PC data for improving computational efficiency. In this paper, we first explore the opportunities of structurizing PC data using Morton code (which is originally designed to map data from a high dimensional space to one dimension, while preserving spatial locality) and observe that there is a huge scope to “skip” the sample and neighbor search computation by operating on the “structurized” PC data. Based on this, we propose two approximation techniques for the sampling and neighbor search stages. We implemented our proposals on an NVIDIA Jetson AGX Xavier edge GPU board. The evaluation results collected on six different workloads show that our design can accelerate the sample and neighbor search stages by 3.68× (up to 5.21×) with minimal impact on inference accuracy. This acceleration in turn results in 1.55× speedup in the end-to-end execution latency and saves 33% of energy expenditure.more » « less
-
null (Ed.)Flat surfaces captured by 3D point clouds are often used for localization, mapping, and modeling. Dense point cloud processing has high computation and memory costs making low-dimensional representations of flat surfaces such as polygons desirable. We present Polylidar3D, a non-convex polygon extraction algorithm which takes as input unorganized 3D point clouds (e.g., LiDAR data), organized point clouds (e.g., range images), or user-provided meshes. Non-convex polygons represent flat surfaces in an environment with interior cutouts representing obstacles or holes. The Polylidar3D front-end transforms input data into a half-edge triangular mesh. This representation provides a common level of abstraction for subsequent back-end processing. The Polylidar3D back-end is composed of four core algorithms: mesh smoothing, dominant plane normal estimation, planar segment extraction, and finally polygon extraction. Polylidar3D is shown to be quite fast, making use of CPU multi-threading and GPU acceleration when available. We demonstrate Polylidar3D’s versatility and speed with real-world datasets including aerial LiDAR point clouds for rooftop mapping, autonomous driving LiDAR point clouds for road surface detection, and RGBD cameras for indoor floor/wall detection. We also evaluate Polylidar3D on a challenging planar segmentation benchmark dataset. Results consistently show excellent speed and accuracy.more » « less
-
With the explosion of intelligent and latency-sensitive applications such as AR/VR, remote health and autonomous driving, mobile edge computing (MEC) has emerged as a promising solution to mitigate the high end-to-end latency of mobile cloud computing (MCC). However, the edge servers have significantly less computing capability compared to the resourceful central cloud. Therefore, a collaborative cloud-edge-local offloading scheme is necessary to accommodate both computationally intensive and latency-sensitive mobile applications. The coexistence of central cloud, edge servers and the mobile device (MD), forming a multi-tiered heterogeneous architecture, makes the optimal application deployment very challenging especially for multi-component applications with component dependencies. This paper addresses the problem of energy and latency efficient application offloading in a collaborative cloud-edgelocal environment. We formulate a multi-objective mixed integer linear program (MILP) with the goal of minimizing the systemwide energy consumption and application end-to-end latency. An approximation algorithm based on LP relaxation and rounding is proposed to address the time complexity. We demonstrate that our approach outperforms existing strategies in terms of application request acceptance ratio, latency and system energy consumption.more » « less
-
Recent breakthroughs in deep learning (DL) have led to the emergence of many intelligent mobile applications and services, but in the meanwhile also pose unprecedented computing challenges on resource-constrained mobile devices. This paper builds a collaborative deep inference system between a resource-constrained mobile device and a powerful edge server, aiming at joining the power of both on-device processing and computation offloading. The basic idea of this system is to partition a deep neural network (DNN) into a front-end part running on the mobile device and a back-end part running on the edge server, with the key challenge being how to locate the optimal partition point to minimize the end-to-end inference delay. Unlike existing efforts on DNN partitioning that rely heavily on a dedicated offline profiling stage to search for the optimal partition point, our system has a built-in online learning module, called Autodidactic Neurosurgeon (ANS), to automatically learn the optimal partition point on-the-fly. Therefore, ANS is able to closely follow the changes of the system environment by generating new knowledge for adaptive decision making. The core of ANS is a novel contextual bandit learning algorithm, called μLinUCB, which not only has provable theoretical learning performance guarantee but also is ultra-lightweight for easy real-world implementation. We implement our system on a video stream object detection testbed to validate the design of ANS and evaluate its performance. The experiments show that ANS significantly outperforms state-of-the-art benchmarks in terms of tracking system changes and reducing the end-to-end inference delay.more » « less
