Search

3D Gaussian Splatting (3DGS) has shown a powerful capability for novel view synthesis due to its detailed expressive ability and highly efficient rendering speed. Unfortunately, creating relightable 3D assets with 3DGS is still problematic, particularly for reflective objects, as its discontinuous representation raises difficulties in constraining geometries. Inspired by previous works, the signed distance field (SDF) can serve as an effective way for geometry regularization. However, a direct incorporation between Gaussians and SDF significantly slows training. To this end, we propose GS-ROR for reflective objects relighting with 3DGS aided by SDF priors. At the core of our method is the mutual supervision of the depth and normal between deferred Gaussians and SDF, which avoids the expensive volume rendering of SDF. Thanks to this mutual supervision, the learned deferred Gaussians are well-constrained with a minimal time cost. As the Gaussians are rendered in a deferred shading mode, while the alpha-blended Gaussians are smooth, individual Gaussians may still be outliers, yielding floater artifacts. Therefore, we further introduce an SDF-aware pruning strategy to remove Gaussian outliers, which are located distant from the surface defined by SDF, avoiding the floater issue. Consequently, our method outperforms the existing Gaussian-based inverse rendering methods in terms of relighting quality. Our method also exhibits competitive relighting quality compared to NeRF-based methods with at most 25% of training time and allows rendering at 200+ frames per second on an RTX4090.

Microstructure uniformity of the Al-Nd target materials with Al11Nd3 significantly affects the performance of the fabricated film, which is widely used as wiring material in largesize thin-film transistor liquid crystal display (TFT-LCD) panels. Understanding the inherent mechanical properties and chemical bonds of Al11Nd3 is crucial for homogenizing the Al-Nd target. Here, by a combined experimental and ab-initio theoretical study, the microstructure and deformability of the Al-3wt%Nd alloy and the inherent mechanical properties and chemical bonds of Al11Nd3 are investigated comprehensively. The Al-3wt%Nd alloy is composed of the pre-eutectic {\alpha}-Al matrix and the eutectic {\alpha}-Al and a high stable {\alpha}-Al11Nd3 phases. During the plastic deformation, the eutectic microstructure transforms from a cellular to a lamellar shape, while the morphology and dimension of {\alpha}-Al11Nd3 are not changed significantly. By examining ideal tensile strength, elastic moduli, hardness and brittleness-ductility, the hardnessbrittleness of {\alpha}-Al11Nd3 is quantitatively evaluated, accounting for its difficulties of plastic deformation and fragmentation. Combining band structure, population analysis, topological analysis and crystal orbital Hamilton population, it is revealed that {\alpha}-Al11Nd3 possesses two types of chemical bonds: the Nd-Al and Al-Al bonds. The former is a typical ionic bond with electron transfer from Nd to Al, while the latter, dominated by both 3s-3p and 3p-3p interactions, is a weak covalent bond. The mixed chemical bond is responsible for the high hardness-brittleness of {\alpha}-Al11Nd3. This work is expected to lay a foundation for Al-Nd alloy and catalyze the fabrication of high-quality Al-Nd target materials., Comment: 11 figures,5 tables

Reconstructive phase transitions involving breaking and reconstruction of primary chemical bonds are ubiquitous and important for many technological applications. In contrast to displacive phase transitions, the dynamics of reconstructive phase transitions are usually slow due to the large energy barrier. Nevertheless, the reconstructive phase transformation from $\beta$- to $\lambda$-Ti$_3$O$_5$ exhibits an ultrafast and reversible behavior. Despite extensive studies, the underlying microscopic mechanism remains unclear. Here, we discover a kinetically favorable in-plane nucleated layer-by-layer transformation mechanism through metadynamics and large-scale molecular dynamics simulations. This is enabled by developing an efficient machine learning potential with near first-principles accuracy through an on-the-fly active learning method and an advanced sampling technique. Our results reveal that the $\beta$-$\lambda$ phase transformation initiates with the formation of two-dimensional nuclei in the $ab$-plane and then proceeds layer-by-layer through a multistep barrier-lowering kinetic process via intermediate metastable phases. Our work not only provides important insight into the ultrafast and reversible nature of the $\beta$-$\lambda$ transition, but also presents useful strategies and methods for tackling other complex structural phase transitions., Comment: 26 pages,23 figures (including Supporting Information)

Recent advancements in diffusion models have showcased their impressive capacity to generate visually striking images. Nevertheless, ensuring a close match between the generated image and the given prompt remains a persistent challenge. In this work, we identify that a crucial factor leading to the text-image mismatch issue is the inadequate cross-modality relation learning between the prompt and the output image. To better align the prompt and image content, we advance the cross-attention with an adaptive mask, which is conditioned on the attention maps and the prompt embeddings, to dynamically adjust the contribution of each text token to the image features. This mechanism explicitly diminishes the ambiguity in semantic information embedding from the text encoder, leading to a boost of text-to-image consistency in the synthesized images. Our method, termed MaskDiffusion, is training-free and hot-pluggable for popular pre-trained diffusion models. When applied to the latent diffusion models, our MaskDiffusion can significantly improve the text-to-image consistency with negligible computation overhead compared to the original diffusion models.

Finite-temperature ductility-brittleness and electronic structures of Al$_3$Sc, Al$_2$Sc and AlSc are studied comparatively by first-principles calculations and ab-initio molecular dynamics. Results show that Al$_3$Sc and Al$_2$Sc are inherently brittle at both ground state and finite temperatures. By contrast, AlSc possesses a significantly superior ductility evaluated from all Pugh's, Pettifor's and Poisson's ductility-brittleness criteria. At ground state, AlSc meets the criteria of ductile according to Pugh's and Poisson's theories, while it is categorized as the brittle in the frame of Pettifor's picture. With the increasing temperature, the ductility of all the studied compounds exhibits a noticeable improvement. In particular, as the temperature rises, the Cauchy pressure of AlSc undergoes a transition from negative to positive. Thus, at high temperatures (T > 600 K), AlSc is unequivocally classified as the ductile from all criteria considered. In all Al$_3$Sc, Al$_2$Sc and AlSc, the Al-Al bond, originated from s-p and p-p orbital hybridizations, and the Al-Sc bond, dominated by p-d covalent hybridization, are the first and second strongest chemical bonds, respectively. To explain the difference in mechanical properties of the studied compounds, the mean bond strength (MBS) is evaluated. The weaker Al-Al bond in AlSc, leading to a smaller MBS, could be the origin for the softer elastic stiffness and superior intrinsic ductility. The longer length of the Al-Al bond in AlSc is responsible for its weaker bond strength. Furthermore, the enhanced metallicity of the Al-Al bond in AlSc would also contribute to its exceptional ductility., Comment: 14 pages, 10 figures

We present All-Pairs Multi-Field Transforms (AMT), a new network architecture for video frame interpolation. It is based on two essential designs. First, we build bidirectional correlation volumes for all pairs of pixels, and use the predicted bilateral flows to retrieve correlations for updating both flows and the interpolated content feature. Second, we derive multiple groups of fine-grained flow fields from one pair of updated coarse flows for performing backward warping on the input frames separately. Combining these two designs enables us to generate promising task-oriented flows and reduce the difficulties in modeling large motions and handling occluded areas during frame interpolation. These qualities promote our model to achieve state-of-the-art performance on various benchmarks with high efficiency. Moreover, our convolution-based model competes favorably compared to Transformer-based models in terms of accuracy and efficiency. Our code is available at https://github.com/MCG-NKU/AMT., Comment: Accepted to CVPR2023

Lighting is a determining factor in photography that affects the style, expression of emotion, and even quality of images. Creating or finding satisfying lighting conditions, in reality, is laborious and time-consuming, so it is of great value to develop a technology to manipulate illumination in an image as post-processing. Although previous works have explored techniques based on the physical viewpoint for relighting images, extensive supervisions and prior knowledge are necessary to generate reasonable images, restricting the generalization ability of these works. In contrast, we take the viewpoint of image-to-image translation and implicitly merge ideas of the conventional physical viewpoint. In this paper, we present an Illumination-Aware Network (IAN) which follows the guidance from hierarchical sampling to progressively relight a scene from a single image with high efficiency. In addition, an Illumination-Aware Residual Block (IARB) is designed to approximate the physical rendering process and to extract precise descriptors of light sources for further manipulations. We also introduce a depth-guided geometry encoder for acquiring valuable geometry- and structure-related representations once the depth information is available. Experimental results show that our proposed method produces better quantitative and qualitative relighting results than previous state-of-the-art methods. The code and models are publicly available on https://github.com/NK-CS-ZZL/IAN., Comment: Accepted for publication as a Regular paper in the IEEE Transactions on Image Processing (T-IP)