ExtrudeNet: Unsupervised Inverse Sketch-and-Extrude for Shape Parsing

Ren, Daxuan; Zheng, Jianmin; Cai, Jianfei; Li, Jiatong; Zhang, Junzhe

doi:10.1007/978-3-031-20086-1_28

Cited by 16 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast to their reliance on 2D labels, SECAD-Net is trained in a self-supervised manner. Most closely related to our work is ExtrudeNet [34]. SECAD-Net distinguishes itself from ExtrudeNet in several significant aspects: i) Following the traditional reconstruction process, ExtrudeNet first predicts the parameters of Bézier curves and then converts them into SDFs.…”

Section: Related Workmentioning

confidence: 99%

“…We thoroughly compare our method with two types of primitive-based CAD reconstruction methods that output editable CAD models, including two CSG-like methods (i.e., UCSG-Net [19], CSG-Stump [33]) and two cylinder decomposition counterpart (i.e., point2Cyl [46]), Extru-deNet [34]. For each method, we adopt the implementation provided by the corresponding authors, and use the same training strategy for training and fine-tuning.…”

Section: Comparison On Cad Reconstructionmentioning

confidence: 99%

See 1 more Smart Citation

SECAD-Net: Self-Supervised CAD Reconstruction by Learning Sketch-Extrude Operations

Li¹,

Guo²,

Zhang³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Comparison On Cad Reconstructionmentioning

confidence: 99%

SECAD-Net: Self-Supervised CAD Reconstruction by Learning Sketch-Extrude Operations

Li¹,

Guo²,

Zhang³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Figure 2.12: Overview of some methods that output sketch and extrude sequences. These figures are taken from their corresponding publications (a) [168], (b) [151], (c) [123].…”

Section: Sketch and Extrudementioning

confidence: 99%

“…Point2Cyl is also considered a representative work in "primitive detection", since it first segments the point cloud into patches, and then reconstruct extrusion cylinders from those patches, as shown in Figure2.12 (b).The works mentioned above all apply supervised training with ground truth command sequences. ExtrudeNet[123] can learn such sequences unsupervisedly. It uses closed cubic…”

mentioning

confidence: 99%

A Review of Deep Learning-Powered Mesh Reconstruction Methods

Feng¹

2023

Preprint

View full text Add to dashboard Cite

With the recent advances in hardware and rendering techniques, 3D models have emerged everywhere in our life. Yet creating 3D shapes is arduous and requires significant professional knowledge. Meanwhile, Deep learning has enabled high-quality 3D shape reconstruction from various sources, making it a viable approach to acquiring 3D shapes with minimal effort. Importantly, to be used in common 3D applications, the reconstructed shapes need to be represented as polygonal meshes, which is a challenge for neural networks due to the irregularity of mesh tessellations.In this survey, we provide a comprehensive review of mesh reconstruction methods that are powered by machine learning. We first describe various representations for 3D shapes in the deep learning context. Then we review the development of 3D mesh reconstruction methods from voxels, point clouds, single images, and multi-view images. Finally, we identify several challenges in this field and propose potential future directions.

show abstract

“…In CSG‐Stump [RZC*21], the out‐put CSG program is a union over intersections of either primitives or complements of primitives. ExtrudeNet [RZC*22] extends upon the CAPRI‐Net architecture by replacing quadric primitives with extruded 2D sketches, while maintaining end‐to‐end differentiability. UCSGNet [KZK20] explores a higher diversity of program structures, by evaluating many boolean operations in parallel, but this comes at the cost of more complex programs with worse reconstruction performance.…”

Section: Application: 3d Shapesmentioning

confidence: 99%

Neurosymbolic Models for Computer Graphics

Ritchie

Guerrero

Jones

et al. 2023

Computer Graphics Forum

View full text Add to dashboard Cite

Procedural models (i.e. symbolic programs that output visual data) are a historically‐popular method for representing graphics content: vegetation, buildings, textures, etc. They offer many advantages: interpretable design parameters, stochastic variations, high‐quality outputs, compact representation, and more. But they also have some limitations, such as the difficulty of authoring a procedural model from scratch. More recently, AI‐based methods, and especially neural networks, have become popular for creating graphic content. These techniques allow users to directly specify desired properties of the artifact they want to create (via examples, constraints, or objectives), while a search, optimization, or learning algorithm takes care of the details. However, this ease of use comes at a cost, as it's often hard to interpret or manipulate these representations. In this state‐of‐the‐art report, we summarize research on neurosymbolic models in computer graphics: methods that combine the strengths of both AI and symbolic programs to represent, generate, and manipulate visual data. We survey recent work applying these techniques to represent 2D shapes, 3D shapes, and materials & textures. Along the way, we situate each prior work in a unified design space for neurosymbolic models, which helps reveal underexplored areas and opportunities for future research.

show abstract

ExtrudeNet: Unsupervised Inverse Sketch-and-Extrude for Shape Parsing

Cited by 16 publications

References 19 publications

SECAD-Net: Self-Supervised CAD Reconstruction by Learning Sketch-Extrude Operations

SECAD-Net: Self-Supervised CAD Reconstruction by Learning Sketch-Extrude Operations

A Review of Deep Learning-Powered Mesh Reconstruction Methods

Neurosymbolic Models for Computer Graphics

Contact Info

Product

Resources

About