BoxNet: A Deep Learning Method for 2D Bounding Box Estimation from Bird's-Eye View Point Cloud

Nezhadarya, Ehsan; Liu, Yang; Liu, Bingbing

doi:10.1109/ivs.2019.8814058

Cited by 11 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Application of pose estimation To demonstrate the utility of accumulated radar points for downstream applications, we apply a pose estimation method, i.e., BoxNet [25], on the accumulated 2D radar points via our full velocity and Table 3: Comparison of pose estimation performance: average error in center and orientation as well as Intersection over Union (IoU), by using BoxNet [25] on radar points accumulated using our velocity and the radial velocity as a baseline.…”

Section: Quantitative Resultsmentioning

confidence: 99%

“…In Fig. 2, each heat map shows point-wise velocity error under different depth ranges, i.e [0, 25), [25,50) and [50, ∞) meters as well as various α ranges, i.e. [0, 30), [30,60) and [60, 90] degrees, where α is the angle between actual moving direction and radial direction of a radar point and ranges from 0 to 90 degrees.…”

Section: Velocity Estimation Error For Different Depths and αmentioning

confidence: 99%

See 1 more Smart Citation

Full-Velocity Radar Returns by Radar-Camera Fusion

Long

Morris

Liu

et al. 2021

Preprint

View full text Add to dashboard Cite

A distinctive feature of Doppler radar is the measurement of velocity in the radial direction for radar points. However, the missing tangential velocity component hampers object velocity estimation as well as temporal integration of radar sweeps in dynamic scenes. Recognizing that fusing camera with radar provides complementary information to radar, in this paper we present a closed-form solution for the point-wise, full-velocity estimate of Doppler returns using the corresponding optical flow from camera images. Additionally, we address the association problem between radar returns and camera images with a neural network that is trained to estimate radar-camera correspondences. Experimental results on the nuScenes dataset verify the validity of the method and show significant improvements over the state-of-the-art in velocity estimation and accumulation of radar points.

show abstract

Section: Quantitative Resultsmentioning

confidence: 99%

Section: Velocity Estimation Error For Different Depths and αmentioning

confidence: 99%

Full-Velocity Radar Returns by Radar-Camera Fusion

Long

Morris

Liu

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In such a way, it makes a great demand on the tooth centroid prediction in our method. Previous detection methods for 3D point clouds [3,13,18] generally use the furthest point sampling (FPS) method to uniformly select the sampling points for generating proposals. For tooth centroid detection, the sampled points by FPS method generally contain irrelevant points, such as one located on the tooth crown and gingiva, which may lead to inaccurate proposals for tooth centroids.…”

Section: Introductionmentioning

confidence: 99%

DArch: Dental Arch Prior-assisted 3D Tooth Instance Segmentation

Qiu¹,

Ye²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

Automatic tooth instance segmentation on 3D dental models is a fundamental task for computer-aided orthodontic treatments. Existing learning-based methods rely heavily on expensive point-wise annotations. To alleviate this problem, we are the first to explore a low-cost annotation way for 3D tooth instance segmentation, i.e., labeling all tooth centroids and only a few teeth for each dental model. Regarding the challenge when only weak annotation is provided, we present a dental arch prior-assisted 3D tooth segmentation method, namely DArch. Our DArch consists of two stages, including tooth centroid detection and tooth instance segmentation. Accurately detecting the tooth centroids can help locate the individual tooth, thus benefiting the segmentation. Thus, our DArch proposes to leverage the dental arch prior to assist the detection. Specifically, we firstly propose a coarse-to-fine method to estimate the dental arch, in which the dental arch is initially generated by Bezier curve regression, and then a graph-based convolutional network (GCN) is trained to refine it. With the estimated dental arch, we then propose a novel Arch-aware Point Sampling (APS) method to assist the tooth centroid proposal generation. Meantime, a segmentor is independently trained using a patch-based training strategy, aiming to segment a tooth instance from a 3D patch centered at the tooth centroid. Experimental results on 4, 773 dental models have shown our DArch can accurately segment each tooth of a dental model, and its performance is superior to the state-of-the-art methods.

show abstract

DArch: Dental Arch Prior-assisted 3D Tooth Instance Segmentation with Weak Annotations

Qiu

Chen

et al. 2022

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

BoxNet: A Deep Learning Method for 2D Bounding Box Estimation from Bird's-Eye View Point Cloud

Cited by 11 publications

References 24 publications

Full-Velocity Radar Returns by Radar-Camera Fusion

Full-Velocity Radar Returns by Radar-Camera Fusion

DArch: Dental Arch Prior-assisted 3D Tooth Instance Segmentation

DArch: Dental Arch Prior-assisted 3D Tooth Instance Segmentation with Weak Annotations

Contact Info

Product

Resources

About