3DIAS: 3D Shape Reconstruction with Implicit Algebraic Surfaces

Yavartanoo, Mohsen; Chung, Jaeyoung; Neshatavar, Reyhaneh; Lee, Kyoung Mu

doi:10.1109/iccv48922.2021.01222

Cited by 10 publications

(10 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some latest researches have investigated decomposing or fitting 3D objects into primitive representations by deep learning approach [15,36,49,58,63,77]. Other methods related to primitives focus on deep shape generation models, where the primitives act as intermediate representations or building blocks [33,70,80].…”

Section: Primitives In 3d Deep Learningmentioning

confidence: 99%

Primitive3D: 3D Object Dataset Synthesis from Randomly Assembled Primitives

Li¹,

Ding²,

Tong³

et al. 2022

Preprint

View full text Add to dashboard Cite

Numerous advancements in deep learning can be attributed to the access to large-scale and well-annotated datasets. However, such a dataset is prohibitively expensive in 3D computer vision due to the substantial collection cost. To alleviate this issue, we propose a cost-effective method for automatically generating a large amount of 3D objects with annotations. In particular, we synthesize objects simply by assembling multiple random primitives. These objects are thus auto-annotated with part labels originating from primitives. This allows us to perform multi-task learning by combining the supervised segmentation with unsupervised reconstruction. Considering the large overhead of learning on the generated dataset, we further propose a dataset distillation strategy to remove redundant samples regarding a target dataset. We conduct extensive experiments for the downstream tasks of 3D object classification. The results indicate that our dataset, together with multitask pretraining on its annotations, achieves the best performance compared to other commonly used datasets. Further study suggests that our strategy can improve the model performance by pretraining and fine-tuning scheme, especially for the dataset with a small scale. In addition, pretraining with the proposed dataset distillation method can save 86% of the pretraining time with negligible performance degradation. We expect that our attempt provides a new datacentric perspective for training 3D deep models.

show abstract

Section: Primitives In 3d Deep Learningmentioning

confidence: 99%

Primitive3D: 3D Object Dataset Synthesis from Randomly Assembled Primitives

Li¹,

Ding²,

Tong³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Decomposing shapes into parts with specific attributes have been extensively studied in computer graphics [61,19,76,95,93]. Recent deep learning based methods tried to resolve this problem by learning primitives using a data-driven strategy [16,68,69,20,76,59,89]. The primitives could be convex polytopes [68,16], 3D Gaussian functions or spheres [22,72].…”

Section: Related Workmentioning

confidence: 99%

Latent Partition Implicit with Surface Codes for 3D Representation

Chen¹,

Liu²,

Han³

2022

Preprint

View full text Add to dashboard Cite

Deep implicit functions have shown remarkable shape modeling ability in various 3D computer vision tasks. One drawback is that it is hard for them to represent a 3D shape as multiple parts. Current solutions learn various primitives and blend the primitives directly in the spatial space, which still struggle to approximate the 3D shape accurately.To resolve this problem, we introduce a novel implicit representation to represent a single 3D shape as a set of parts in the latent space, towards both highly accurate and plausibly interpretable shape modeling. Our insight here is that both the part learning and the part blending can be conducted much easier in the latent space than in the spatial space. We name our method Latent Partition Implicit (LPI), because of its ability of casting the global shape modeling into multiple local part modeling, which partitions the global shape unity. LPI represents a shape as Signed Distance Functions (SDFs) using surface codes. Each surface code is a latent code representing a part whose center is on the surface, which enables us to flexibly employ intrinsic attributes of shapes or additional surface properties. Eventually, LPI can reconstruct both the shape and the parts on the shape, both of which are plausible meshes. LPI is a multilevel representation, which can partition a shape into different numbers of parts after training. LPI can be learned without ground truth signed distances, point normals or any supervision for part partition. LPI outperforms the latest methods under the widely used benchmarks in terms of reconstruction accuracy and modeling interpretability. Our code, data and models are available at https://github.com/chenchao15/LPI.

show abstract

“…Marching Cubes [43]. A noteworthy branch of work builds hy-brid implicit/explicit representations [14,20,52,87] based mostly on differentiable space partitioning.…”

Section: Related Workmentioning

confidence: 99%

“…The level-set derived from these MLPs can be visualized through techniques like ray marching [32] or tessellated into an explicit shape using methods like Marching Cubes [52]. Another notable line of research involves the development of hybrid implicit/explicit representations [20,23,62,94], primarily based on differentiable space partitioning. To simultaneously represent collections of shapes, implicit neural models necessitate conditioning mechanisms.…”

Section: Related Workmentioning

confidence: 99%

Few ‘Zero Level Set’-Shot Learning of Shape Signed Distance Functions in Feature Space

Ouasfi

Boukhayma

2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

While current state-of-the-art generalizable implicit neural shape models [7,54] rely on the inductive bias of convolutions, it is still not entirely clear how properties emerging from such biases are compatible with the task of 3D reconstruction from point cloud. We explore an alternative approach to generalizability in this context. We relax the intrinsic model bias (i.e. using MLPs to encode local features as opposed to convolutions) and constrain the hypothesis space instead with an auxiliary regularization related to the reconstruction task, i.e. denoising. The resulting model is the first only-MLP locally conditioned implicit shape reconstruction from point cloud network with fast feed forward inference. Point cloud borne features and denoising offsets are predicted from an exclusively MLP-made network in a single forward pass. A decoder predicts occupancy probabilities for queries anywhere in space by pooling nearby features from the point cloud order-invariantly, guided by denoised relative positional encoding. We outperform the state-of-the-art convolutional method [7] while using half the number of model parameters.

show abstract

3DIAS: 3D Shape Reconstruction with Implicit Algebraic Surfaces

Cited by 10 publications

References 19 publications

Primitive3D: 3D Object Dataset Synthesis from Randomly Assembled Primitives

Primitive3D: 3D Object Dataset Synthesis from Randomly Assembled Primitives

Latent Partition Implicit with Surface Codes for 3D Representation

Few ‘Zero Level Set’-Shot Learning of Shape Signed Distance Functions in Feature Space

Contact Info

Product

Resources

About