MB-Net: MergeBoxes for Real-Time 3D Vehicles Detection

Gählert, Nils; Mayer, M; Schneider, Lukas; Franke, Uwe; Denzler, Joachim

doi:10.1109/ivs.2018.8500395

Cited by 13 publications

(9 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This innovation enables InvPT++ to markedly surpass InvPT while utilizing 22.51% fewer FLOPs in the transformer decoder. We remarkably extend our experiments to investigate the effectiveness of InvPT++ by considering two new and more challenging 2D/3D multi-task scene understanding benchmarks, i.e., the Cityscapes [19] and Cityscapes-3D [20] datasets. Substantial results and analysis in qualitative and quantitative aspects are further provided in the experiments.…”

Section: Global Spatial Interaction Simultaneous All-task Interactionmentioning

confidence: 99%

See 1 more Smart Citation

Inverted Pyramid Multi-task Transformer for Dense Scene Understanding

Hanrong

2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Multi-task scene understanding aims to design models that can simultaneously predict several scene understanding tasks with one versatile model. Previous studies typically process multi-task features in a more local way, and thus cannot effectively learn spatially global and cross-task interactions, which hampers the models' ability to fully leverage the consistency of various tasks in multitask learning. To tackle this problem, we propose an Inverted Pyramid multi-task Transformer, capable of modeling cross-task interaction among spatial features of different tasks in a global context. Specifically, we first utilize a transformer encoder to capture task-generic features for all tasks. And then, we design a transformer decoder to establish spatial and cross-task interaction globally, and a novel UP-Transformer block is devised to increase the resolutions of multi-task features gradually and establish cross-task interaction at different scales. Furthermore, two types of Cross-Scale Self-Attention modules, i.e., Fusion Attention and Selective Attention, are proposed to efficiently facilitate cross-task interaction across different feature scales. An Encoder Feature Aggregation strategy is further introduced to better model multi-scale information in the decoder. Comprehensive experiments on several 2D/3D multi-task benchmarks clearly demonstrate our proposal's effectiveness, establishing significant state-of-the-art performances. The codes are publicly available here.

show abstract

Section: Global Spatial Interaction Simultaneous All-task Interactionmentioning

confidence: 99%

“…Fig. 7: Qualitative comparison between our predictions and the ground-truth labels (GT) for joint 2D-3D multi-task scene understanding using the Cityscapes-3D dataset [20].…”

Section: Selective Attentionmentioning

confidence: 99%

Inverted Pyramid Multi-task Transformer for Dense Scene Understanding

Hanrong

2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…The projection of the predicted 3D BBox should be consistent with the predicted 2D BBox. This is utilized to build geometric constraints in [27,14,26] to determine the depth. Thanks to the promising performance of convolutional neural networks in 2D object detection, more and more approaches [3,20,31,30,29,24,12,16] have been proposed to directly predict 3D BBoxes using well-designed convolutional neural network for monocular 3D object detection.…”

Section: Related Workmentioning

confidence: 99%

M3DSSD: Monocular 3D Single Stage Object Detector

Luo

Dai

Shao

et al. 2021

2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

In this paper, we propose a Monocular 3D Single Stage object Detector (M3DSSD) with feature alignment and asymmetric non-local attention. Current anchor-based monocular 3D object detection methods suffer from feature mismatching. To overcome this, we propose a twostep feature alignment approach. In the first step, the shape alignment is performed to enable the receptive field of the feature map to focus on the pre-defined anchors with high confidence scores. In the second step, the center alignment is used to align the features at 2D/3D centers. Further, it is often difficult to learn global information and capture long-range relationships, which are important for the depth prediction of objects. Therefore, we propose a novel asymmetric non-local attention block with multiscale sampling to extract depth-wise features. The proposed M3DSSD achieves significantly better performance than the monocular 3D object detection methods on the KITTI dataset, in both 3D object detection and bird's eye view tasks. The code is released at https://github. com/mumianyuxin/M3DSSD.

show abstract

“… Deeplearning: Voxelnet [49]; Second [50]; MVX-Net [51]  Other: Particle filter [52]  Original 3D data information can be retained  SSD [64]; YOLOv3 [65]; MB-Net [66]; EZ-Net [67];…”

Section: Volumetricmentioning

confidence: 99%

A Survey on Sensor Technologies for Unmanned Ground Vehicles

Liu

Yuan

2020

2020 3rd International Conference on Unmanned Systems (ICUS)

View full text Add to dashboard Cite

Unmanned ground vehicles (UGVs) have a huge development potential in both civilian and military fields, and have become the focus of research in various countries. In addition, high-precision, high-reliability sensors are significant for UGVs' efficient operation. This paper proposes a brief review on sensor technologies for UGVs. Firstly, characteristics of various sensors are introduced. Then the strengths and weaknesses of different sensors as well as their application scenarios are compared. Furthermore, sensor applications in some existing UGVs are summarized. Finally, the hotspots of sensor technologies are forecasted to point the development direction.

show abstract

MB-Net: MergeBoxes for Real-Time 3D Vehicles Detection

Cited by 13 publications

References 24 publications

Inverted Pyramid Multi-task Transformer for Dense Scene Understanding

Inverted Pyramid Multi-task Transformer for Dense Scene Understanding

M3DSSD: Monocular 3D Single Stage Object Detector

A Survey on Sensor Technologies for Unmanned Ground Vehicles

Contact Info

Product

Resources

About