NeRF-Texture: Synthesizing Neural Radiance Field Textures.

Texture synthesis is a fundamental problem in computer graphics that would benefit various applications. Existing methods are effective in handling 2D image textures. In contrast, many real-world textures contain meso-structure in the 3D geometry space, such as grass, leaves, and fabrics, which cannot be effectively modeled using only 2D image textures. We propose a novel texture synthesis method with Neural Radiance Fields (NeRF) to capture and synthesize textures from given multi-view images. In the proposed NeRF texture representation, a scene with fine geometric details is disentangled into the meso-structure textures and the underlying base shape. This allows textures with meso-structure to be effectively learned as latent features situated on the base shape, which are fed into a NeRF decoder trained simultaneously to represent the rich view-dependent appearance. Using this implicit representation, we can synthesize NeRF-based textures through patch matching of latent features. However, inconsistencies between the metrics of the reconstructed content space and the latent feature space may compromise the synthesis quality. To enhance matching performance, we further regularize the distribution of latent features by incorporating a clustering constraint. In addition to generating NeRF textures over a planar domain, our method can also synthesize NeRF textures over curved surfaces, which are practically useful. Experimental results and evaluations demonstrate the effectiveness of our approach.

Snowflake Point Deconvolution for Point Cloud Completion and Generation With Skip-Transformer.

Most existing point cloud completion methods suffer from the discrete nature of point clouds and the unstructured prediction of points in local regions, which makes it difficult to reveal fine local geometric details. To resolve this issue, we propose SnowflakeNet with snowflake point deconvolution (SPD) to generate complete point clouds. SPD models the generation of point clouds as the snowflake-like growth of points, where child points are generated progressively by splitting their parent points after each SPD. Our insight into the detailed geometry is to introduce a skip-transformer in the SPD to learn the point splitting patterns that can best fit the local regions. The skip-transformer leverages attention mechanism to summarize the splitting patterns used in the previous SPD layer to produce the splitting in the current layer. The locally compact and structured point clouds generated by SPD precisely reveal the structural characteristics of the 3D shape in local patches, which enables us to predict highly detailed geometries. Moreover, since SPD is a general operation that is not limited to completion, we explore its applications in other generative tasks, including point cloud auto-encoding, generation, single image reconstruction, and upsampling. Our experimental results outperform state-of-the-art methods under widely used benchmarks.

PMP-Net++: Point Cloud Completion by Transformer-Enhanced Multi-Step Point Moving Paths.

Point cloud completion concerns to predict missing part for incomplete 3D shapes. A common strategy is to generate complete shape according to incomplete input. However, unordered nature of point clouds will degrade generation of high-quality 3D shapes, as detailed topology and structure of unordered points are hard to be captured during the generative process using an extracted latent code. We address this problem by formulating completion as point cloud deformation process. Specifically, we design a novel neural network, named PMP-Net++, to mimic behavior of an earth mover. It moves each point of incomplete input to obtain a complete point cloud, where total distance of point moving paths (PMPs) should be the shortest. Therefore, PMP-Net++ predicts unique PMP for each point according to constraint of point moving distances. The network learns a strict and unique correspondence on point-level, and thus improves quality of predicted complete shape. Moreover, since moving points heavily relies on per-point features learned by network, we further introduce a transformer-enhanced representation learning network, which significantly improves completion performance of PMP-Net++. We conduct comprehensive experiments in shape completion, and further explore application on point cloud up-sampling, which demonstrate non-trivial improvement of PMP-Net++ over state-of-the-art point cloud completion/up-sampling methods.

An endometrial receptivity scoring system basing on the endometrial thickness, volume, echo, peristalsis, and blood flow evaluated by ultrasonography.

Background: Establishing a successful pregnancy depends on the endometrium and the embryo. It is estimated that suboptimal endometrial receptivity account for one-third of implantation failures. Despite the indepth understanding of the processes associated with embryo-endometrial cross-talk, little progress has been achieved for diagnosis and treatments for suboptimal endometrial receptivity. Methods: This retrospective study included women undergoing their first frozen-thawed embryo transfer (FET) cycles at our reproductive medicine center from March 2021 to August 2021. Transvaginal three-dimensional (3D) ultrasound was performed in the morning on the day of embryo transfer for all the thawed embryo transfer patients, to evaluate endometrial receptivity, including endometrial thickness, echogenicity, volume, movement and blood flow. Results: A total number of 562 patients of FET with 315 pregnancies (56.0%) was analyzed. It was found that only the echo of the endometrial central line was different between the pregnant group and non-pregnant group. Other parameters, such as endometrial thickness, volume, endometrial peristalsis, or the endometrial blood flow were not statistically different between the two groups. Then, according to the relationship between the different groups and the clinical pregnancy rate, a score of 0 to 2 was respectively scored. The sum of the scores for the six items was the patient's endometrial receptivity score. It showed that the clinical pregnancy rate increased as the endometrial receptivity score increased, and when the receptivity score reaches at least 5, the clinical pregnancy rate is significantly improved (63.7% versus 49.5%, P=0.001). Conclusion: We developed an endometrial receptivity scoring system and demonstrated its validity. It may aid clinicians in choosing the useful marker in clinical practice and for informing further research.

High-Quality Textured 3D Shape Reconstruction with Cascaded Fully Convolutional Networks.

We present a learning-based approach to reconstructing high-resolution three-dimensional (3D) shapes with detailed geometry and high-fidelity textures. Albeit extensively studied, algorithms for 3D reconstruction from multi-view depth-and-color (RGB-D) scans are still prone to measurement noise and occlusions; limited scanning or capturing angles also often lead to incomplete reconstructions. Propelled by recent advances in 3D deep learning techniques, in this paper, we introduce a novel computation- and memory-efficient cascaded 3D convolutional network architecture, which learns to reconstruct implicit surface representations as well as the corresponding color information from noisy and imperfect RGB-D maps. The proposed 3D neural network performs reconstruction in a progressive and coarse-to-fine manner, achieving unprecedented output resolution and fidelity. Meanwhile, an algorithm for end-to-end training of the proposed cascaded structure is developed. We further introduce Human10, a newly created dataset containing both detailed and textured full-body reconstructions as well as corresponding raw RGB-D scans of 10 subjects. Qualitative and quantitative experimental results on both synthetic and real-world datasets demonstrate that the presented approach outperforms existing state-of-the-art work regarding visual quality and accuracy of reconstructed models.

Active Exploration of Large 3D Model Repositories.

With broader availability of large-scale 3D model repositories, the need for efficient and effective exploration becomes more and more urgent. Existing model retrieval techniques do not scale well with the size of the database since often a large number of very similar objects are returned for a query, and the possibilities to refine the search are quite limited. We propose an interactive approach where the user feeds an active learning procedure by labeling either entire models or parts of them as "like" or "dislike" such that the system can automatically update an active set of recommended models. To provide an intuitive user interface, candidate models are presented based on their estimated relevance for the current query. From the methodological point of view, our main contribution is to exploit not only the similarity between a query and the database models but also the similarities among the database models themselves. We achieve this by an offline pre-processing stage, where global and local shape descriptors are computed for each model and a sparse distance metric is derived that can be evaluated efficiently even for very large databases. We demonstrate the effectiveness of our method by interactively exploring a repository containing over 100 K models.