ASFM-Net: Asymmetrical Siamese Feature Matching Network for Point Completion

Xia, Yaqi; Xia, Yan; Li, Wei; Song, Rui; Cao, Kailang; Stilla, Uwe

doi:10.1145/3474085.3475348

Cited by 43 publications

(6 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yuan et al [46] designed PCN, which proposed a coarse-to-fine method based on the PointNet [25] and FoldingNet [44], but its decoder often fails to recover rare geometries of objects such as seat backs with gaps, etc. Therefore, after PCN, many other methods [11,27,32,40] focus on multi-step point cloud generation, which is helpful to recover a final point cloud with fine-grained details. Furthermore, following DGCNN [33], some researchers developed graph-based methods [36,37,54] which consider regional geometric details.…”

Section: Related Workmentioning

confidence: 99%

ProxyFormer: Proxy Alignment Assisted Point Cloud Completion with Missing Part Sensitive Transformer

Li¹,

Gao²,

Tan³

et al. 2023

Preprint

View full text Add to dashboard Cite

Problems such as equipment defects or limited viewpoints will lead the captured point clouds to be incomplete. Therefore, recovering the complete point clouds from the partial ones plays an vital role in many practical tasks, and one of the keys lies in the prediction of the missing part. In this paper, we propose a novel point cloud completion approach namely ProxyFormer that divides point clouds into existing (input) and missing (to be predicted) parts and each part communicates information through its proxies. Specifically, we fuse information into point proxy via feature and position extractor, and generate features for missing point proxies from the features of existing point proxies. Then, in order to better perceive the position of missing points, we design a missing part sensitive transformer, which converts random normal distribution into reasonable position information, and uses proxy alignment to refine the missing proxies. It makes the predicted point proxies more sensitive to the features and positions of the missing part, and thus make these proxies more suitable for subsequent coarse-to-fine processes. Experimental results show that our method outperforms state-of-the-art completion networks on several benchmark datasets and has the fastest inference speed. Code is available at https://github. com/I2-Multimedia-Lab/ProxyFormer.

show abstract

Section: Related Workmentioning

confidence: 99%

ProxyFormer: Proxy Alignment Assisted Point Cloud Completion with Missing Part Sensitive Transformer

Li¹,

Gao²,

Tan³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…We concatenate them together and sum up with the resampled partial cloud to move each point of it, where geometric details are reconstructed distinctly. Inspired by ASFM-net [44], we adopt an iterative refinement process that regards a refine point on the first refine stage as a coarse point, and refine it again to get more reasonable results.…”

Section: F I G U R Ementioning

confidence: 99%

Point completion by a Stack‐Style Folding Network with multi‐scaled graphical features

Rao,

Xu,

Zeng

et al. 2023

IET Computer Vision

View full text Add to dashboard Cite

Point cloud completion is prevalent due to the insufficient results from current point cloud acquisition equipments, where a large number of point data failed to represent a relatively complete shape. Existing point cloud completion algorithms, mostly encoder‐decoder structures with grids transform (also presented as folding operation), can hardly obtain a persuasive representation of input clouds due to the issue that their bottleneck‐shape result cannot tell a precise relationship between the global and local structures. For this reason, this article proposes a novel point cloud completion model based on a Stack‐Style Folding Network (SSFN). Firstly, to enhance the deep latent feature extraction, SSFN enhances the exploitation of shape feature extractor by integrating both low‐level point feature and high‐level graphical feature. Next, a precise presentation is obtained from a high dimensional semantic space to improve the reconstruction ability. Finally, a refining module is designed to make a more evenly distributed result. Experimental results shows that our SSFN produces the most promising results of multiple representative metrics with a smaller scale parameters than current models.

show abstract

“…For instance, RFNet (Huang et al 2021), PMP-Net (Wen et al 2021) and PMP-Net++ (Wen et al 2022) completed the points level by level, where the recurrent neural network was utilized to reserve useful information of previous level. ASFM-Net (Xia et al 2021) and SnowflakeNet (Xiang et al 2021) upsampled and moved the points at each refinement iteration. Recently, a two-path network (Zhao et al 2021b) for pairwise completion was proposed to separately complete objects which bear a strong spatial relation.…”

Section: Related Workmentioning

confidence: 99%

CasFusionNet: A Cascaded Network for Point Cloud Semantic Scene Completion by Dense Feature Fusion

Tang

et al. 2023

AAAI

View full text Add to dashboard Cite

Semantic scene completion (SSC) aims to complete a partial 3D scene and predict its semantics simultaneously. Most existing works adopt the voxel representations, thus suffering from the growth of memory and computation cost as the voxel resolution increases. Though a few works attempt to solve SSC from the perspective of 3D point clouds, they have not fully exploited the correlation and complementarity between the two tasks of scene completion and semantic segmentation. In our work, we present CasFusionNet, a novel cascaded network for point cloud semantic scene completion by dense feature fusion. Specifically, we design (i) a global completion module (GCM) to produce an upsampled and completed but coarse point set, (ii) a semantic segmentation module (SSM) to predict the per-point semantic labels of the completed points generated by GCM, and (iii) a local refinement module (LRM) to further refine the coarse completed points and the associated labels from a local perspective. We organize the above three modules via dense feature fusion in each level, and cascade a total of four levels, where we also employ feature fusion between each level for sufficient information usage. Both quantitative and qualitative results on our compiled two point-based datasets validate the effectiveness and superiority of our CasFusionNet compared to state-of-the-art methods in terms of both scene completion and semantic segmentation. The codes and datasets are available at: https://github.com/JinfengX/CasFusionNet.

show abstract

ASFM-Net: Asymmetrical Siamese Feature Matching Network for Point Completion

Cited by 43 publications

References 61 publications

ProxyFormer: Proxy Alignment Assisted Point Cloud Completion with Missing Part Sensitive Transformer

ProxyFormer: Proxy Alignment Assisted Point Cloud Completion with Missing Part Sensitive Transformer

Point completion by a Stack‐Style Folding Network with multi‐scaled graphical features

CasFusionNet: A Cascaded Network for Point Cloud Semantic Scene Completion by Dense Feature Fusion

Contact Info

Product

Resources

About