arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1703.04079cs.CVcs.CG
收藏

SurfNet:使用深度残差网络生成3D形状曲面

SurfNet: Generating 3D shape surfaces using deep residual networks

MIT(麻省理工学院);IIT Kanpur(印度理工学院坎普尔分校);UT Austin(得克萨斯大学奥斯汀分校);Purdue(普渡大学)

Ayan Sinha, Asim Unmesh, Qixing Huang, Karthik Ramani

摘要:

3D形状模型使用由形成表面的多边形组成的顶点和面\ie自然地参数化。然而,使用卷积神经网络的预测和生成任务的当前3D学习范例集中于对象的体素化表示。将卷积算子从传统的2D提升到3D会导致高计算开销,而几乎没有额外的好处,因为大多数几何信息都包含在表面边界上。在这里,我们研究的问题,直接产生的刚性和非刚性形状的三维形状表面使用深卷积神经网络。我们开发了一个程序来创建一致的“几何图像”表示一类3D对象的形状表面。然后,我们使用这种一致的表示为类别特定的形状表面生成的参数表示或图像,通过开发新的扩展深度残差网络的几何图像生成的任务。我们的实验表明,我们的网络学习了形状表面的有意义的表示,允许它在形状方向和姿势之间插值,发明新的形状表面,并从以前看不见的图像重建3D形状表面。

英文摘要:

3D shape models are naturally parameterized using vertices and faces, \ie, composed of polygons forming a surface. However, current 3D learning paradigms for predictive and generative tasks using convolutional neural networks focus on a voxelized representation of the object. Lifting convolution operators from the traditional 2D to 3D results in high computational overhead with little additional benefit as most of the geometry information is contained on the surface boundary. Here we study the problem of directly generating the 3D shape surface of rigid and non-rigid shapes using deep convolutional neural networks. We develop a procedure to create consistent `geometry images' representing the shape surface of a category of 3D objects. We then use this consistent representation for category-specific shape surface generation from a parametric representation or an image by developing novel extensions of deep residual networks for the task of geometry image generation. Our experiments indicate that our network learns a meaningful representation of shape surfaces allowing it to interpolate between shape orientations and poses, invent new shape surfaces and reconstruct 3D shape surfaces from previously unseen images.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑