16 results
Search Results
Now showing 1 - 10 of 16
Item Attention And Positional Encoding Are (Almost) All You Need For Shape Matching(The Eurographics Association and John Wiley & Sons Ltd., 2023) Raganato, Alessandro; Pasi, Gabriella; Melzi, Simone; Memari, Pooran; Solomon, JustinThe fast development of novel approaches derived from the Transformers architecture has led to outstanding performance in different scenarios, from Natural Language Processing to Computer Vision. Recently, they achieved impressive results even in the challenging task of non-rigid shape matching. However, little is known about the capability of the Transformer-encoder architecture for the shape matching task, and its performances still remained largely unexplored. In this paper, we step back and investigate the contribution made by the Transformer-encoder architecture compared to its more recent alternatives, focusing on why and how it works on this specific task. Thanks to the versatility of our implementation, we can harness the bi-directional structure of the correspondence problem, making it more interpretable. Furthermore, we prove that positional encodings are essential for processing unordered point clouds. Through a comprehensive set of experiments, we find that attention and positional encoding are (almost) all you need for shape matching. The simple Transformer-encoder architecture, coupled with relative position encoding in the attention mechanism, is able to obtain strong improvements, reaching the current state-of-the-art.Item Factored Neural Representation for Scene Understanding(The Eurographics Association and John Wiley & Sons Ltd., 2023) Wong, Yu-Shiang; Mitra, Niloy J.; Memari, Pooran; Solomon, JustinA long-standing goal in scene understanding is to obtain interpretable and editable representations that can be directly constructed from a raw monocular RGB-D video, without requiring specialized hardware setup or priors. The problem is significantly more challenging in the presence of multiple moving and/or deforming objects. Traditional methods have approached the setup with a mix of simplifications, scene priors, pretrained templates, or known deformation models. The advent of neural representations, especially neural implicit representations and radiance fields, opens the possibility of end-to-end optimization to collectively capture geometry, appearance, and object motion. However, current approaches produce global scene encoding, assume multiview capture with limited or no motion in the scenes, and do not facilitate easy manipulation beyond novel view synthesis. In this work, we introduce a factored neural scene representation that can directly be learned from a monocular RGB-D video to produce object-level neural presentations with an explicit encoding of object movement (e.g., rigid trajectory) and/or deformations (e.g., nonrigid movement). We evaluate ours against a set of neural approaches on both synthetic and real data to demonstrate that the representation is efficient, interpretable, and editable (e.g., change object trajectory). Code and data are available at: http://geometry.cs.ucl.ac.uk/projects/2023/factorednerf/.Item Cross-Shape Attention for Part Segmentation of 3D Point Clouds(The Eurographics Association and John Wiley & Sons Ltd., 2023) Loizou, Marios; Garg, Siddhant; Petrov, Dmitry; Averkiou, Melinos; Kalogerakis, Evangelos; Memari, Pooran; Solomon, JustinWe present a deep learning method that propagates point-wise feature representations across shapes within a collection for the purpose of 3D shape segmentation. We propose a cross-shape attention mechanism to enable interactions between a shape's point-wise features and those of other shapes. The mechanism assesses both the degree of interaction between points and also mediates feature propagation across shapes, improving the accuracy and consistency of the resulting point-wise feature representations for shape segmentation. Our method also proposes a shape retrieval measure to select suitable shapes for crossshape attention operations for each test shape. Our experiments demonstrate that our approach yields state-of-the-art results in the popular PartNet dataset.Item 3D Keypoint Estimation Using Implicit Representation Learning(The Eurographics Association and John Wiley & Sons Ltd., 2023) Zhu, Xiangyu; Du, Dong; Huang, Haibin; Ma, Chongyang; Han, Xiaoguang; Memari, Pooran; Solomon, JustinIn this paper, we tackle the challenging problem of 3D keypoint estimation of general objects using a novel implicit representation. Previous works have demonstrated promising results for keypoint prediction through direct coordinate regression or heatmap-based inference. However, these methods are commonly studied for specific subjects, such as human bodies and faces, which possess fixed keypoint structures. They also suffer in several practical scenarios where explicit or complete geometry is not given, including images and partial point clouds. Inspired by the recent success of advanced implicit representation in reconstruction tasks, we explore the idea of using an implicit field to represent keypoints. Specifically, our key idea is employing spheres to represent 3D keypoints, thereby enabling the learnability of the corresponding signed distance field. Explicit keypoints can be extracted subsequently by our algorithm based on the Hough transform. Quantitative and qualitative evaluations also show the superiority of our representation in terms of prediction accuracy.Item Singularity-Free Frame Fields for Line Drawing Vectorization(The Eurographics Association and John Wiley & Sons Ltd., 2023) Guțan, Olga; Hegde, Shreya; Berumen, Erick Jimenez; Bessmeltsev, Mikhail; Chien, Edward; Memari, Pooran; Solomon, JustinState-of-the-art methods for line drawing vectorization rely on generated frame fields for robust direction disambiguation, with each of the two axes aligning to different intersecting curve tangents around junctions. However, a common source of topological error for such methods are frame field singularities. To remedy this, we introduce the first frame field optimization framework guaranteed to produce singularity-free fields aligned to a line drawing. We first perform a convex solve for a roughly-aligned orthogonal frame field (cross field), and then comb away its internal singularities with an optimal transport–based matching. The resulting topology of the field is strictly maintained with the machinery of discrete trivial connections in a final, non-convex optimization that allows non-orthogonality of the field, improving smoothness and tangent alignment. Our frame fields can serve as a drop-in replacement for frame field optimizations used in previous work, improving the quality of the final vectorizations.Item PowerRTF: Power Diagram based Restricted Tangent Face for Surface Remeshing(The Eurographics Association and John Wiley & Sons Ltd., 2023) Yao, Yuyou; Liu, Jingjing; Fei, Yue; Wu, Wenming; Zhang, Gaofeng; Yan, Dong-Ming; Zheng, Liping; Memari, Pooran; Solomon, JustinTriangular meshes of superior quality are important for geometric processing in practical applications. Existing approximative CVT-based remeshing methodology uses planar polygonal facets to fit the original surface, simplifying the computational complexity. However, they usually do not consider surface curvature. Topological errors and outliers can also occur in the close sheet surface remeshing, resulting in wrong meshes. With this regard, we present a novel method named PowerRTF, an extension of the restricted tangent face (RTF) in conjunction with the power diagram, to better approximate the original surface with curvature adaption. The idea is to introduce a weight property to each sample point and compute the power diagram on the tangent face to produce area-controlled polygonal facets. Based on this, we impose the variable-capacity constraint and centroid constraint to the PowerRTF, providing the trade-off between mesh quality and computational efficiency. Moreover, we apply a normal verification-based inverse side point culling method to address the topological errors and outliers in close sheet surface remeshing. Our method independently computes and optimizes the PowerRTF per sample point, which is efficiently implemented in parallel on the GPU. Experimental results demonstrate the effectiveness, flexibility, and efficiency of our method.Item HalfedgeCNN for Native and Flexible Deep Learning on Triangle Meshes(The Eurographics Association and John Wiley & Sons Ltd., 2023) Ludwig, Ingmar; Tyson, Daniel; Campen, Marcel; Memari, Pooran; Solomon, JustinWe describe HalfedgeCNN, a collection of modules to build neural networks that operate on triangle meshes. Taking inspiration from the (edge-based) MeshCNN, convolution, pooling, and unpooling layers are consistently defined on the basis of halfedges of the mesh, pairs of oppositely oriented virtual instances of each edge. This provides benefits over alternative definitions on the basis of vertices, edges, or faces. Additional interface layers enable support for feature data associated with such mesh entities in input and output as well. Due to being defined natively on mesh entities and their neighborhoods, lossy resampling or interpolation techniques (to enable the application of operators adopted from image domains) do not need to be employed. The operators have various degrees of freedom that can be exploited to adapt to application-specific needs.Item Feature-Preserving Offset Mesh Generation from Topology-Adapted Octrees(The Eurographics Association and John Wiley & Sons Ltd., 2023) Zint, Daniel; Maruani, Nissim; Rouxel-Labbé, Mael; Alliez, Pierre; Memari, Pooran; Solomon, JustinWe introduce a reliable method to generate offset meshes from input triangle meshes or triangle soups. Our method proceeds in two steps. The first step performs a Dual Contouring method on the offset surface, operating on an adaptive octree that is refined in areas where the offset topology is complex. Our approach substantially reduces memory consumption and runtime compared to isosurfacing methods operating on uniform grids. The second step improves the output Dual Contouring mesh with an offset-aware remeshing algorithm to reduce the normal deviation between the mesh facets and the exact offset. This remeshing process reconstructs concave sharp features and approximates smooth shapes in convex areas up to a user-defined precision. We show the effectiveness and versatility of our method by applying it to a wide range of input meshes. We also benchmark our method on the Thingi10k dataset: watertight and topologically 2-manifold offset meshes are obtained for 100% of the cases.Item Lightweight Curvature Estimation on Point Clouds with Randomized Corrected Curvature Measures(The Eurographics Association and John Wiley & Sons Ltd., 2023) Lachaud, Jacques-Olivier; Coeurjolly, David; Labart, Céline; Romon, Pascal; Thibert, Boris; Memari, Pooran; Solomon, JustinThe estimation of differential quantities on oriented point cloud is a classical step for many geometry processing tasks in computer graphics and vision. Even if many solutions exist to estimate such quantities, they usually fail at satisfying both a stable estimation with theoretical guarantee, and the efficiency of the associated algorithm. Relying on the notion of corrected curvature measures [LRT22, LRTC20] designed for surfaces, the method introduced in this paper meets both requirements. Given a point of interest and a few nearest neighbours, our method estimates the whole curvature tensor information by generating random triangles within these neighbours and normalising the corrected curvature measures by the corrected area measure. We provide a stability theorem showing that our pointwise curvatures are accurate and convergent, provided the noise in position and normal information has a variance smaller than the radius of neighbourhood. Experiments and comparisons with the state-of-the-art confirm that our approach is more accurate and much faster than alternatives. The method is fully parallelizable, requires only one nearest neighbour request per point of computation, and is trivial to implement.Item Deep Deformation Detail Synthesis for Thin Shell Models(The Eurographics Association and John Wiley & Sons Ltd., 2023) Chen, Lan; Gao, Lin; Yang, Jie; Xu, Shibiao; Ye, Juntao; Zhang, Xiaopeng; Lai, Yu-Kun; Memari, Pooran; Solomon, JustinIn physics-based cloth animation, rich folds and detailed wrinkles are achieved at the cost of expensive computational resources and huge labor tuning. Data-driven techniques make efforts to reduce the computation significantly by utilizing a preprocessed database. One type of methods relies on human poses to synthesize fitted garments, but these methods cannot be applied to general cloth animations. Another type of methods adds details to the coarse meshes obtained through simulation, which does not have such restrictions. However, existing works usually utilize coordinate-based representations which cannot cope with large-scale deformation, and requires dense vertex correspondences between coarse and fine meshes. Moreover, as such methods only add details, they require coarse meshes to be sufficiently close to fine meshes, which can be either impossible, or require unrealistic constraints to be applied when generating fine meshes. To address these challenges, we develop a temporally and spatially as-consistent-as-possible deformation representation (named TS-ACAP) and design a DeformTransformer network to learn the mapping from low-resolution meshes to ones with fine details. This TS-ACAP representation is designed to ensure both spatial and temporal consistency for sequential large-scale deformations from cloth animations. With this TS-ACAP representation, our DeformTransformer network first utilizes two mesh-based encoders to extract the coarse and fine features using shared convolutional kernels, respectively. To transduct the coarse features to the fine ones, we leverage the spatial and temporal Transformer network that consists of vertex-level and frame-level attention mechanisms to ensure detail enhancement and temporal coherence of the prediction. Experimental results show that our method is able to produce reliable and realistic animations in various datasets at high frame rates with superior detail synthesis abilities compared to existing methods.