Object as Hotspots: An Anchor-Free 3D Object Detection Approach via Firing of Hotspots

Chen, Qi; Sun, Lin; Wang, Zhixin; Jia, Kui; Yuille, Alan

doi:10.1007/978-3-030-58589-1_5

Cited by 130 publications

(63 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[62], [63] utilize multiple detection heads while [24] explores the object part locations for improving the performance. In addition, [64], [65] predicts bounding box parameters following the anchor-free paradigm. These grid-based methods are generally efficient for accurate 3D proposal generation but the receptive fields are constraint by the kernel size of 2D/3D convolutions.…”

Section: D Object Detection With Point Cloudsmentioning

confidence: 99%

PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection

Shi

Guo²,

Jiang³

et al. 2020

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

1,603

1,045

View full text Add to dashboard Cite

3D object detection is receiving increasing attention from both industry and academia thanks to its wide applications in various fields. In this paper, we propose the Point-Voxel Region based Convolution Neural Networks (PV-RCNNs) for accurate 3D detection from point clouds. First, we propose a novel 3D object detector, PV-RCNN-v1, which employs the voxel-to-keypoint scene encoding and keypoint-to-grid RoI feature abstraction two novel steps. These two steps deeply incorporate both 3D voxel CNN and PointNet-based set abstraction for learning discriminative point-cloud features. Second, we propose a more advanced framework, PV-RCNN-v2, for more efficient and accurate 3D detection. It consists of two major improvements, where the first one is the sectorized proposal-centric strategy for efficiently producing more representative and uniformly distributed keypoints, and the second one is the VectorPool aggregation to replace set abstraction for better aggregating local point-cloud features with much less resource consumption. With these two major modifications, our PV-RCNN-v2 runs more than twice as fast as the v1 version while still achieving better performance on the large-scale Waymo Open Dataset with 150m × 150m detection range. Extensive experiments demonstrate that our proposed PV-RCNNs significantly outperform previous state-of-the-art 3D detection methods on both the Waymo Open Dataset and the highly-competitive KITTI benchmark.

show abstract

Section: D Object Detection With Point Cloudsmentioning

confidence: 99%

PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection

Shi

Guo²,

Jiang³

et al. 2020

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

1,603

1,045

View full text Add to dashboard Cite

show abstract

“…GNNs reuse graph edges in all layers and avoid repetitive grouping and sampling of points. Table 1 is in terms of the Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI) benchmark, point-GNN achieved high accuracy using only point clouds and exhibited accuracy similar to that of fusion-based algorithms [4,[15][16][17][18][19][20][21][22][23][24][25].…”

Section: Point Cloud-based 3d Object Detectionmentioning

confidence: 99%

Point-Graph Neural Network Based Novel Visual Positioning System for Indoor Navigation

et al. 2021

View full text Add to dashboard Cite

Indoor localization is a basic element in location-based services (LBSs), including seamless indoor and outdoor navigation, location-based precision marketing, spatial recognition in robotics, augmented reality, and mixed reality. The popularity of LBSs in the augmented reality and mixed reality fields has increased the demand for a stable and efficient indoor positioning method. However, the problem of indoor visual localization has not been appropriately addressed, owing to the strict trade-off between accuracy and cost. Therefore, we use point cloud and RGB characteristic information for the accurate acquisition of three-dimensional indoor space. The proposed method is a novel visual positioning system (VPS) capable of determining the user’s position by matching the pose information of the object estimated by the improved point-graph neural network (GNN) with the pose information label of a voxel database object addressed in predefined voxel units. We evaluated the performance of the proposed system considering a stationary object in indoor space. The results verify that high positioning accuracy and direction estimation can be efficiently achieved. Thus, spatial information of indoor space estimated using the proposed novel VPS can aid in indoor navigation.

show abstract

“…Anchor free methods: Nowadays, anchor-free methods [40]- [44] attracted much attention because they achieve excellent performance in the object detection task. In contrast to anchor-based methods, anchor-free methods don't need to set up a large number of hand-designed anchors.…”

Section: A Object Detectionmentioning

confidence: 99%

AFOD: An Adaptable Framework for Object Detection in Event-based Vision

Zhang¹,

Wang²

2021

Preprint

View full text Add to dashboard Cite

<div>Event-based vision is a novel bio-inspired vision that has attracted the interest of many researchers. As a neuromorphic vision, the sensor is different from the traditional frame-based cameras. It has such advantages that conventional frame-based cameras can’t match, e.g., high temporal resolution, high dynamic range(HDR), sparse and minimal motion blur. Recently, a lot of computer vision approaches have been proposed with demonstrated success. However, there is a lack of some general methods to expand the scope of the application of event-based vision. To be able to effectively bridge the gap between conventional computer vision and event-based vision, in this paper, we propose an adaptable framework for object detection in event-based vision.</div>

show abstract

Object as Hotspots: An Anchor-Free 3D Object Detection Approach via Firing of Hotspots

Cited by 130 publications

References 40 publications

PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection

PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection

Point-Graph Neural Network Based Novel Visual Positioning System for Indoor Navigation

AFOD: An Adaptable Framework for Object Detection in Event-based Vision

Contact Info

Product

Resources

About