We introduce TM-NET, a novel deep generative model for synthesizing textured meshes in a part-aware manner. Once trained, the network can generate novel textured meshes from scratch or predict textures for a given 3D mesh, without image guidance. Plausible and diverse textures can be generated for the same mesh part, while texture compatibility between parts in the same shape is achieved via conditional generation. Specifically, our method produces texture maps for individual shape parts, each as a deformable box, leading to a natural UV map with limited distortion. The network separately embeds part geometry (via a PartVAE) and part texture (via a TextureVAE) into their respective latent spaces, so as to facilitate learning texture probability distributions conditioned on geometry. We introduce a conditional autoregressive model for texture generation, which can be conditioned on both part geometry and textures already generated for other parts to achieve texture compatibility. To produce high-frequency texture details, our Tex-tureVAE operates in a high-dimensional latent space via dictionary-based vector quantization. We also exploit transparencies in the texture as an effective means to model complex shape structures including topological details. Extensive experiments demonstrate the plausibility, quality, and diversity of the textures and geometries generated by our network, while avoiding inconsistency issues that are common to novel view synthesis methods. CCS Concepts: • Computing methodologies → Shape modeling.

No abstract

In this work, we tackle the problem of single imagebased 3D shape retrieval (IBSR), where we seek to find the most matched shape of a given single 2D image from a shape repository. Most of the existing works learn to embed 2D images and 3D shapes into a common feature space and perform metric learning using a triplet loss. Inspired by the great success in recent contrastive learning works on self-supervised representation learning, we propose a novel IBSR pipeline leveraging contrastive learning. We note that adopting such cross-modal contrastive learning between 2D images and 3D shapes into IBSR tasks is non-trivial and challenging: contrastive learning requires very strong data augmentation in constructed positive pairs to learn the feature invariance, whereas traditional metric learning works do not have this requirement. Moreover, object shape and appearance are entangled in 2D query images, thus making the learning task more difficult than contrasting single-modal data. To mitigate the challenges, we propose to use multi-view grayscale rendered images from the 3D shapes as a shape representation. We then introduce a strong data augmentation technique based on color transfer, which can significantly but naturally change the appearance of the query image, effectively satisfying the need for contrastive learning. Finally, we propose to incorporate a novel category-level contrastive loss that helps distinguish similar objects from different categories, in addition to classic instance-level contrastive loss. Our experiments demonstrate that our approach achieves the best performance on * Corresponding Author is Lin Gao all the three popular IBSR benchmarks, including Pix3D, Stanford Cars, and Comp Cars, outperforming the previous state-of-the-art from 4% -15% on retrieval accuracy.

This study implemented the measurement results and administrative information obtained from the hole plate into the Digital Calibration Certificate (DCC). The DCC comprises three parts: Norms and Standards, Hierarchical Structure, and XML as Exchange Format. DCCs play a significant role in the field of metrology and statistics by ensuring data interoperability, correctness, and traceability during the conversion and transmission process. The hole plate is a length standard used for two-dimensional geometric error measurements. We evaluated the accuracy of the high-precision coordinate measuring machine (CMM) in measuring a hole plate and compared the measurement error results obtained from the hole plate with those of the laser interferometer, autocollimator, and angle square. The results show that the maximum difference in linear error is −0.30 μm, the maximum difference in angle error is −0.78″, and the maximum difference in squareness error is 4.54″. The XML is designed for machine-readability and is modeled and edited using the XMLSpy 2022 software, which is based on information published by PTB. The administrative management and measurement results tasks are presented in PDF format, which is designed for human-readability and ease of use. Overall, we implemented the measurement results and information obtained from the hole plate into the DCC.

Recently, a four-axis coordinate measuring machine, which consists of three linear axes and a single rotary axis, has been more widely used than a traditional three-axis coordinate measuring machine. The volumetric error influences the accuracy of the four-axis coordinate measuring machine. 27 parametric errors contribute to the volumetric error. This study proposes a new methodology to analyze the parametric and volumetric error of the four-axis coordinate measuring machine using a hole plate. First, the hole plate setup sequentially in three different planes. We measured the hole plate by using five different styluses. Second, 27 parametric errors were analyzed using the coordinate deviations. The volumetric error was constructed using the homogeneous transform matrixes. The volumetric error ranges were from 0.35 µm to 1.55 µm and from 0.35 µm to 2.83 µm without and with the single rotary axis. Third, three commercial instruments, namely a laser interferometer, an autocollimator, and a polygon-autocollimator, were used to validate the proposed methodology to verify the measured parametric errors. The absolute maximum differences, compared with the laser interferometer for three parametric positioning errors and the autocollimator for six parametric rotational errors for the three linear axes, were 0.56 µm and 0.54", respectively. Moreover, the absolute maximum difference of one parametric positioning error for the single rotary axis, comparing with the polygon-autocollimator, was 0.75". The En-values were 0.27, 0.54, and 0.27, respectively. The results demonstrate the proposed methodology's effectiveness and reliability to the industry's four-axis coordinate measuring machines.