LC-NAS: Latency Constrained Neural Architecture Search for Point Cloud Networks

Li, Guohao; Xu, Mengmeng; Giancola, Silvio; Thabet, Ali; Ghanem, Bernard

doi:10.48550/arxiv.2008.10309

Cited by 4 publications

(6 citation statements)

References 45 publications

(91 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also works on 3D shape classification [27,19], but their overall frameworks do not exceed that set by [25]. [47,20] is closer to our work, in the sense that it uses NAS to optimize for segmentation and detection on 3D scenes (KITTI [13]). But generalizing the terminology used in [23], we believe there is also a two-level hierarchy in 3D neural architecture designs, with the outer macro-level controlling the views of the data / features, and the inner micro-level being the specifics of the neural layers.…”

Section: Neural Architecture Searchmentioning

confidence: 86%

LidarNAS: Unifying and Searching Neural Architectures for 3D Point Clouds

Liu¹,

Leng²,

Sun³

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Developing neural models that accurately understand objects in 3D point clouds is essential for the success of robotics and autonomous driving. However, arguably due to the higher-dimensional nature of the data (as compared to images), existing neural architectures exhibit a large variety in their designs, including but not limited to the views considered, the format of the neural features, and the neural operations used. Lack of a unified framework and interpretation makes it hard to put these designs in perspective, as well as systematically explore new ones. In this paper, we begin by proposing a unified framework of such, with the key idea being factorizing the neural networks into a series of view transforms and neural layers. We demonstrate that this modular framework can reproduce a variety of existing works while allowing a fair comparison of backbone designs. Then, we show how this framework can easily materialize into a concrete neural architecture search (NAS) space, allowing a principled NAS-for-3D exploration. In performing evolutionary NAS on the 3D object detection task on the Waymo Open Dataset, not only do we outperform the state-of-the-art models, but also report the interesting finding that NAS tends to discover the same macro-level architecture concept for both the vehicle and pedestrian classes.

show abstract

Section: Neural Architecture Searchmentioning

confidence: 86%

LidarNAS: Unifying and Searching Neural Architectures for 3D Point Clouds

Liu¹,

Leng²,

Sun³

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…For point-based methods, SGAS [169] explores a DARTS [187]like network topology with graph convolution layers as candidate operations in the search space. Built upon SGAS, LC-NAS [170] further incorporates the hardware feedback into the pipeline by training a differentiable latency regressor to predict the network latency on target platforms. The predicted latency acts as a regularization term and soft constraint during training.…”

Section: Efficient Point Cloud Processingmentioning

confidence: 99%

Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications

Cai

Lin

et al. 2022

ACM Trans. Des. Autom. Electron. Syst.

View full text Add to dashboard Cite

Deep neural networks (DNNs) have achieved unprecedented success in the field of artificial intelligence (AI), including computer vision, natural language processing, and speech recognition. However, their superior performance comes at the considerable cost of computational complexity, which greatly hinders their applications in many resource-constrained devices, such as mobile phones and Internet of Things (IoT) devices. Therefore, methods and techniques that are able to lift the efficiency bottleneck while preserving the high accuracy of DNNs are in great demand to enable numerous edge AI applications. This article provides an overview of efficient deep learning methods, systems, and applications. We start from introducing popular model compression methods, including pruning, factorization, quantization, as well as compact model design. To reduce the large design cost of these manual solutions, we discuss the AutoML framework for each of them, such as neural architecture search (NAS) and automated pruning and quantization. We then cover efficient on-device training to enable user customization based on the local data on mobile devices. Apart from general acceleration techniques, we also showcase several task-specific accelerations for point cloud, video, and natural language processing by exploiting their spatial sparsity and temporal/token redundancy. Finally, to support all these algorithmic advancements, we introduce the efficient deep learning system design from both software and hardware perspectives.

show abstract

“…Here, #P denotes the number of parameters, #M denotes the number of MACs, and L denotes the measured latency (on a single NVIDIA GTX 1080 Ti GPU). * : numbers are from Li et al [86], which are measured on a single NVIDIA RTX 2080 GPU.…”

Section: Shapenetmentioning

confidence: 99%

“…We compare PVCNN with state-of-the-art pointbased methods including PointNet [3], PointNet++ [4], Deep LPN [83], SpiderCNN [18], PointCNN [6] and ResGCN [84]. We also compare PVNAS with automatically-designed point cloud segmentation networks including SGAS [85] and LC-NAS [86]. To ensure fair comparisons, we follow the original experiment setting in Mo et al [78] to train separate models for different object classes.…”

Section: Partnetmentioning

confidence: 99%

PVNAS: 3D Neural Architecture Search with Point-Voxel Convolution

Liu¹,

Tang²,

Zhao³

et al. 2021

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

3D neural networks are widely used in real-world applications (e.g., AR/VR headsets, self-driving cars). They are required to be fast and accurate; however, limited hardware resources on edge devices make these requirements rather challenging. Previous work processes 3D data using either voxel-based or point-based neural networks, but both types of 3D models are not hardware-efficient due to the large memory footprint and random memory access. In this paper, we study 3D deep learning from the efficiency perspective. We first systematically analyze the bottlenecks of previous 3D methods. We then combine the best from point-based and voxel-based models together and propose a novel hardware-efficient 3D primitive, Point-Voxel Convolution (PVConv). We further enhance this primitive with the sparse convolution to make it more effective in processing large (outdoor) scenes. Based on our designed 3D primitive, we introduce 3D Neural Architecture Search (3D-NAS) to explore the best 3D network architecture given a resource constraint. We evaluate our proposed method on six representative benchmark datasets, achieving state-of-the-art performance with 1.8-23.7× measured speedup. Furthermore, our method has been deployed to the autonomous racing vehicle of MIT Driverless, achieving larger detection range, higher accuracy and lower latency.

show abstract

LC-NAS: Latency Constrained Neural Architecture Search for Point Cloud Networks

Cited by 4 publications

References 45 publications

LidarNAS: Unifying and Searching Neural Architectures for 3D Point Clouds

LidarNAS: Unifying and Searching Neural Architectures for 3D Point Clouds

Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications

PVNAS: 3D Neural Architecture Search with Point-Voxel Convolution

Contact Info

Product

Resources

About