A Two-Stage Pillar Feature-Encoding Network for Pillar-Based 3D Object Detection

Xu, Huiling; Xiang, Dong; Wu, Wenxuan; Yu, Biao; Zhu, Hui

doi:10.3390/wevj14060146

Cited by 4 publications

(1 citation statement)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Owing to the powerful feature extraction capabilities of CNNs, object detection for autonomous driving has made significant breakthroughs [17][18][19][20][21][22][23]. Ren et al [17] proposed Faster R-CNN, which used a Region Proposal Network (RPN) instead of Selective Search to generate Region of Interest (RoI) proposals faster.…”

Section: Tracking By Detectionmentioning

confidence: 99%

Online Multiple Object Tracking Using Min-Cost Flow on Temporal Window for Autonomous Driving

Wei,

Huang,

Zhang

et al. 2023

WEVJ

View full text Add to dashboard Cite

Multiple object tracking (MOT), as a core technology for environment perception in autonomous driving, has attracted attention from researchers. Combing the advantages of batch global optimization, we present a novel online MOT framework for autonomous driving, consisting of feature extraction and data association on a temporal window. In the feature extraction stage, we design a three-channel appearance feature extraction network based on metric learning by using ResNet50 as the backbone network and the triplet loss function and employ a Kalman Filter with a constant acceleration motion model to optimize and predict the object bounding box information, so as to obtain reliable and discriminative object representation features. For data association, to reduce the ID switches, the min-cost flow of global association is introduced within the temporal window composed of consecutive multi-frame images. The trajectories within the temporal window are divided into two categories, active trajectories and inactive trajectories, and the appearance, motion affinities between each category of trajectories, and detections are calculated, respectively. Based on this, a sparse affinity network is constructed, and the data association is achieved using the min-cost flow problem of the network. Qualitative experimental results on KITTI MOT public benchmark dataset and real-world campus scenario sequences validate the effectiveness and robustness of our method. Compared with the homogeneous, vision-based MOT methods, quantitative experimental results demonstrate that our method has competitive advantages in terms of higher order tracking accuracy, association accuracy, and ID switches.

show abstract

Section: Tracking By Detectionmentioning

confidence: 99%

Online Multiple Object Tracking Using Min-Cost Flow on Temporal Window for Autonomous Driving

Wei,

Huang,

Zhang

et al. 2023

WEVJ

View full text Add to dashboard Cite

show abstract

PillarBAPI: Enhancing Pillar-Based 3D Object Detection through attentive Pseudo-Image Feature Extraction

Wang,

Yu,

Cheng

et al. 2024

Preprint

View full text Add to dashboard Cite

Three-dimensional (3D) object detection plays a pivotal role in autonomous driving and intelligent robots. However, current methods often struggle with false and missing detections, especially for small objects. To address these challenges, this paper introduces PillarBAPI, a high-performance 3D object detection network that improves pillar feature coding and enhances point cloud feature representation. PillarBAPI proposes an Attention-based Point and Pillar Feature Extraction (APFE) module to reduce information loss from maximum pooling and enable the model to focus on local and global features. Additionally, we introduce a Pseudo-Image Feature Extraction Network (PIFE) and a novel neck design, B-ASPP, to enhance pseudo-image feature extraction and promote multiscale feature fusion. Extensive experiments on the KITTI dataset demonstrate that Pillar BAPI achieves significant improvements in both 3D and Bird's Eye View (BEV) benchmarks, particularly for small target detection. The contributions of this work lie in the enhanced pillar feature coding, attentive feature extraction, and efficient multi-scale feature fusion, collectively contributing to improved 3D object detection performance. The code will be released at https://github.com/wangjie16138/PillarBAPI/.

show abstract

Real-Time Multimodal 3D Object Detection with Transformers

Liu,

Duan

2024

WEVJ

View full text Add to dashboard Cite

The accuracy and real-time performance of 3D object detection are key factors limiting its widespread application. While cameras capture detailed color and texture features, they lack depth information compared to LiDAR. Multimodal detection combining both can improve results but incurs significant computational overhead, affecting real-time performance. To address these challenges, this paper presents a real-time multimodal fusion model called Fast Transfusion that combines the benefits of LiDAR and camera sensors and reduces the computational burden of their fusion. Specifically, our Fast Transfusion method uses QConv (Quick Convolution) to replace the convolutional backbones compared to other models. QConv concentrates the convolution operations at the feature map center, where the most information resides, to expedite inference. It also utilizes deformable convolution to better match the actual shapes of detected objects, enhancing accuracy. And the model incorporates EH Decoder (Efficient and Hybrid Decoder) which decouples multiscale fusion into intra-scale interaction and cross-scale fusion, efficiently decoding and integrating features extracted from multimodal data. Furthermore, our proposed semi-dynamic query selection refines the initialization of object queries. On the KITTI 3D object detection dataset, our proposed approach reduced the inference time by 36 ms and improved 3D AP by 1.81% compared to state-of-the-art methods.

show abstract

A Two-Stage Pillar Feature-Encoding Network for Pillar-Based 3D Object Detection

Cited by 4 publications

References 44 publications

Online Multiple Object Tracking Using Min-Cost Flow on Temporal Window for Autonomous Driving

Online Multiple Object Tracking Using Min-Cost Flow on Temporal Window for Autonomous Driving

PillarBAPI: Enhancing Pillar-Based 3D Object Detection through attentive Pseudo-Image Feature Extraction

Real-Time Multimodal 3D Object Detection with Transformers

Contact Info

Product

Resources

About