Vladimir G. Kim

Data-driven shape analysis and processing

Data-driven methods serve an increasingly important role in discovering geometric, structural, and semantic relationships between shapes. In contrast to traditional approaches that process shapes in isolation of each other, data-driven methods aggregate information from 3D model collections to improve the analysis, modeling and editing of shapes. Through reviewing the literature, we provide an overview of the main concepts and components of these methods, as well as discuss their application to classification, segmentation, matching, reconstruction, modeling and exploration, as well as scene analysis and synthesis. We conclude our report with ideas that can inspire future research in data-driven shape analysis and processing.

Entropic metric alignment for correspondence problems

Many shape and image processing tools rely on computation of correspondences between geometric domains. Efficient methods that stably extract soft matches in the presence of diverse geometric structures have proven to be valuable for shape retrieval and transfer of labels or semantic information. With these applications in mind, we present an algorithm for probabilistic correspondence that optimizes an entropy-regularized Gromov-Wasserstein (GW) objective. Built upon recent developments in numerical optimal transportation, our algorithm is compact, provably convergent, and applicable to any geometric domain expressible as a metric measure matrix. We provide comprehensive experiments illustrating the convergence and applicability of our algorithm to a variety of graphics tasks. Furthermore, we expand entropic GW correspondence to a framework for other matching problems, incorporating partial distance matrices, user guidance, shape exploration, symmetry detection, and joint analysis of more than two domains. These applications expand the scope of entropic GW correspondence to major shape analysis problems and are stable to distortion and noise.

Convolutional neural networks on surfaces via seamless toric covers

The recent success of convolutional neural networks (CNNs) for image processing tasks is inspiring research efforts attempting to achieve similar success for geometric tasks. One of the main challenges in applying CNNs to surfaces is defining a natural convolution operator on surfaces. In this paper we present a method for applying deep learning to sphere-type shapes using a global seamless parameterization to a planar flat-torus, for which the convolution operator is well defined. As a result, the standard deep learning framework can be readily applied for learning semantic, high-level properties of the shape. An indication of our success in bridging the gap between images and surfaces is the fact that our algorithm succeeds in learning semantic information from an input of raw low-dimensional feature vectors. We demonstrate the usefulness of our approach by presenting two applications: human body segmentation, and automatic landmark detection on anatomical surfaces. We show that our algorithm compares favorably with competing geometric deep-learning algorithms for segmentation tasks, and is able to produce meaningful correspondences on anatomical surfaces where hand-crafted features are bound to fail.

Physics-driven pattern adjustment for direct 3D garment editing

Designers frequently reuse existing designs as a starting point for creating new garments. In order to apply garment modifications, which the designer envisions in 3D, existing tools require meticulous manual editing of 2D patterns. These 2D edits need to account both for the envisioned geometric changes in the 3D shape, as well as for various physical factors that affect the look of the draped garment. We propose a new framework that allows designers to directly apply the changes they envision in 3D space; and creates the 2D patterns that replicate this envisioned target geometry when lifted into 3D via a physical draping simulation. Our framework removes the need for laborious and knowledge-intensive manual 2D edits and allows users to effortlessly mix existing garment designs as well as adjust for garment length and fit. Following each user specified editing operation we first compute a target 3D garment shape, one that maximally preserves the input garments style-its proportions, fit and shape-subject to the modifications specified by the user. We then automatically compute 2D patterns that recreate the target garment shape when draped around the input mannequin within a user-selected simulation environment. To generate these patterns, we propose a fixed-point optimization scheme that compensates for the deformation due to the physical forces affecting the drape and is independent of the underlying simulation tool used. Our experiments show that this method quickly and reliably converges to patterns that, under simulation, form the desired target look, and works well with different black-box physical simulators. We demonstrate a range of edited and resimulated garments, and further validate our approach via expert and amateur critique, and comparisons to alternative solutions.

Learning Local Shape Descriptors from Part Correspondences with Multiview Convolutional Networks

We present a new local descriptor for 3D shapes, directly applicable to a wide range of shape analysis problems such as point correspondences, semantic segmentation, affordance prediction, and shape-to-scan matching. The descriptor is produced by a convolutional network that is trained to embed geometrically and semantically similar points close to one another in descriptor space. The network processes surface neighborhoods around points on a shape that are captured at multiple scales by a succession of progressively zoomed-out views, taken from carefully selected camera positions. We leverage two extremely large sources of data to train our network. First, since our network processes rendered views in the form of 2D images, we repurpose architectures pretrained on massive image datasets. Second, we automatically generate a synthetic dense point correspondence dataset by nonrigid alignment of corresponding shape parts in a large collection of segmented 3D models. As a result of these design choices, our network effectively encodes multiscale local context and fine-grained surface detail. Our network can be trained to produce either category-specific descriptors or more generic descriptors by learning from multiple shape categories. Once trained, at test time, the network extracts local descriptors for shapes without requiring any part segmentation as input. Our method can produce effective local descriptors even for shapes whose category is unknown or different from the ones used while training. We demonstrate through several experiments that our learned local descriptors are more discriminative compared to state-of-the-art alternatives and are effective in a variety of shape analysis applications.

Learning hierarchical shape segmentation and labeling from online repositories

We propose a method for converting geometric shapes into hierarchically segmented parts with part labels. Our key idea is to train category-specific models from the scene graphs and part names that accompany 3D shapes in public repositories. These freely-available annotations represent an enormous, untapped source of information on geometry. However, because the models and corresponding scene graphs are created by a wide range of modelers with different levels of expertise, modeling tools, and objectives, these models have very inconsistent segmentations and hierarchies with sparse and noisy textual tags. Our method involves two analysis steps. First, we perform a joint optimization to simultaneously cluster and label parts in the database while also inferring a canonical tag dictionary and part hierarchy. We then use this labeled data to train a method for hierarchical segmentation and labeling of new 3D shapes. We demonstrate that our method can mine complex information, detecting hierarchies in man-made objects and their constituent parts, obtaining finer scale details than existing alternatives. We also show that, by performing domain transfer using a few supervised examples, our technique outperforms fully-supervised techniques that require hundreds of manually-labeled models.

Complementme: weakly-supervised component suggestions for 3D modeling

Assembly-based tools provide a powerful modeling paradigm for non-expert shape designers. However, choosing a component from a large shape repository and aligning it to a partial assembly can become a daunting task. In this paper we describe novel neural network architectures for suggesting complementary components and their placement for an incomplete 3D part assembly. Unlike most existing techniques, our networks are trained on unlabeled data obtained from public online repositories, and do not rely on consistent part segmentations or labels. Absence of labels poses a challenge in indexing the database of parts for the retrieval. We address it by jointly training embedding and retrieval networks, where the first indexes parts by mapping them to a low-dimensional feature space, and the second maps partial assemblies to appropriate complements. The combinatorial nature of part arrangements poses another challenge, since the retrieval network is not a function: several complements can be appropriate for the same input. Thus, instead of predicting a single output, we train our network to predict a probability distribution over the space of part embeddings. This allows our method to deal with ambiguities and naturally enables a UI that seamlessly integrates user preferences into the design process. We demonstrate that our method can be used to design complex shapes with minimal or no user input. To evaluate our approach we develop a novel benchmark for component suggestion systems demonstrating significant improvement over state-of-the-art techniques.

Tooncap: a layered deformable model for capturing poses from cartoon characters

Characters in traditional artwork such as childrens books or cartoon animations are typically drawn once, in fixed poses, with little opportunity to change the characters appearance or re-use them in a different animation. To enable these applications one can fit a consistent parametric deformable model--- a puppet ---to different images of a character, thus establishing consistent segmentation, dense semantic correspondence, and deformation parameters across poses. In this work, we argue that a layered deformable puppet is a natural representation for hand-drawn characters, providing an effective way to deal with the articulation, expressive deformation, and occlusion that are common to this style of artwork. Our main contribution is an automatic pipeline for fitting these models to unlabeled images depicting the same character in various poses. We demonstrate that the output of our pipeline can be used directly for editing and re-targeting animations.

Real-Time Hair Rendering Using Sequential Adversarial Networks

We present an adversarial network for rendering photorealistic hair as an alternative to conventional computer graphics pipelines. Our deep learning approach does not require low-level parameter tuning nor ad-hoc asset design. Our method simply takes a strand-based 3D hair model as input and provides intuitive user-control for color and lighting through reference images. To handle the diversity of hairstyles and its appearance complexity, we disentangle hair structure, color, and illumination properties using a sequential GAN architecture and a semi-supervised training approach. We also introduce an intermediate edge activation map to orientation field conversion step to ensure a successful CG-to-photoreal transition, while preserving the hair structures of the original input data. As we only require a feed-forward pass through the network, our rendering performs in real-time. We demonstrate the synthesis of photorealistic hair images on a wide range of intricate hairstyles and compare our technique with state-of-the-art hair rendering methods.

Learning A Stroke-Based Representation for Fonts: Learning A Stroke-Based Representation for Fonts

Designing fonts and typefaces is a difficult process for both beginner and expert typographers. Existing workflows require the designer to create every glyph, while adhering to many loosely defined design suggestions to achieve an aesthetically appealing and coherent character set. This process can be significantly simplified by exploiting the similar structure character glyphs present across different fonts and the shared stylistic elements within the same font. To capture these correlations, we propose learning a stroke‐based font representation from a collection of existing typefaces. To enable this, we develop a stroke‐based geometric model for glyphs, a fitting procedure to reparametrize arbitrary fonts to our representation. We demonstrate the effectiveness of our model through a manifold learning technique that estimates a low‐dimensional font space. Our representation captures a wide range of everyday fonts with topological variations and naturally handles discrete and continuous variations, such as presence and absence of stylistic elements as well as slants and weights. We show that our learned representation can be used for iteratively improving fit quality, as well as exploratory style applications such as completing a font from a subset of observed glyphs, interpolating or adding and removing stylistic elements in existing fonts.