We introduce a modeling tool which can evolve a set of 3D objects in a functionality-aware manner. Our goal is for the evolution to generate large and diverse sets of plausible 3D objects for data augmentation, constrained modeling, as well as open-ended exploration to possibly inspire new designs. Starting with an initial population of 3D objects belonging to one or more functional categories, we evolve the shapes through part recombination to produce generations of hybrids or crossbreeds between parents from the heterogeneous shape collection. Evolutionary selection of offsprings is guided both by a functional plausibility score derived from functionality analysis of shapes in the initial population and user preference, as in a design gallery. Since cross-category hybridization may result in offsprings not belonging to any of the known functional categories, we develop a means for functionality partial matching to evaluate functional plausibility on partial shapes. We show a variety of plausible hybrid shapes generated by our functionality-aware model evolution, which can complement existing datasets as training data and boost the performance of contemporary data-driven segmentation schemes, especially in challenging cases. Our tool supports constrained modeling, allowing users to restrict or steer the model evolution with functionality labels. At the same time, unexpected yet functional object prototypes can emerge during open-ended exploration owing to structure breaking when evolving a heterogeneous collection.

Scalable sensor simulation is an important yet challenging open problem for safety-critical domains such as self-driving. Current works in image simulation either fail to be photorealistic or do not model the 3D environment and the dynamic objects within, losing high-level control and physical realism. In this paper, we present GeoSim, a geometry-aware image composition process which synthesizes novel urban driving scenarios by augmenting existing images with dynamic objects extracted from other scenes and rendered at novel poses. Towards this goal, we first build a diverse bank of 3D objects with both realistic geometry and appearance from sensor data. During simulation, we perform a novel geometry-aware simulation-by-composition procedure which 1) proposes plausible and realistic object placements into a given scene, 2) renders novel views of dynamic objects from the asset bank, and 3) composes and blends the rendered image segments. The resulting synthetic images are realistic, traffic-aware, and geometrically consistent, allowing our approach to scale to complex use cases. We demonstrate two such important applications: long-range realistic video simulation across multiple camera sensors, and synthetic data generation for data augmentation on downstream segmentation tasks. Please check https://tmux.top/publication/geosim/ for high-resolution video results.

Constructing and animating humans is an important component for building virtual worlds in a wide variety of applications such as virtual reality or robotics testing in simulation. As there are exponentially many variations of humans with different shape, pose and clothing, it is critical to develop methods that can automatically reconstruct and animate humans at scale from real world data. Towards this goal, we represent the pedestrian’s shape, pose and skinning weights as neural implicit functions that are directly learned from data. This representation enables us to handle a wide variety of different pedestrian shapes and poses without explicitly fitting a human parametric body model, allowing us to handle a wider range of human geometries and topologies. We demonstrate the effectiveness of our approach on various datasets and show that our reconstructions outperform existing state-of-the-art methods. Furthermore, our re-animation experiments show that we can generate 3D human animations at scale from a single RGB image (and/or an optional LiDAR sweep) as input.

An intelligent agent operating in the real-world must balance achieving its goal with maintaining the safety and comfort of not only itself, but also other participants within the surrounding scene. This requires jointly reasoning about the behavior of other actors while deciding its own actions as these two processes are inherently intertwined – a vehicle will yield to us if we decide to proceed first at the intersection but will proceed first if we decide to yield. However, this is not captured in most self-driving pipelines, where planning follows prediction. In this paper we propose a novel data-driven, reactive planning objective which allows a self-driving vehicle to jointly reason about its own plans as well as how other actors will react to them. We formulate the problem as an energy-based deep structured model that is learned from observational data and encodes both the planning and prediction problems. Through simulations based on both real-world driving and synthetically generated dense traffic, we demonstrate that our reactive model outperforms a non-reactive variant in successfully completing highly complex maneuvers (lane merges/turns in traffic) faster, without trading off collision rate. Please see our supplementary document https://tinyurl.com/3nukpn5b for all additional details.

Modern self-driving systems heavily rely on deep learning. As a consequence, their performance is influenced significantly by the quality and richness of the training data. Data collection platforms can generate many hours of raw data on a daily basis, however, it is not feasible to label everything. Therefore, it is critical to have a mechanism to identify "what to label". Active learning approaches identify examples to label, but their interestingness is tied to a fixed model performing a particular task. These assumptions are not valid in self-driving, where we must solve a diverse set of tasks (i.e., perception, motion forecasting, and planning) and models frequently evolve over time. In this paper, we introduce a novel approach to dataset selection that exploits a diverse set of criteria that quantize interestingness of traffic scenes. Our experiments on a wide range of tasks and models demonstrate that the proposed curation pipeline is able to select datasets that lead to better generalization and improved performance.

Reliable markerless motion tracking of people participating in a complex group activity from multiple moving cameras is challenging due to frequent occlusions, strong viewpoint and appearance variations, and asynchronous video streams. To solve this problem, reliable association of the same person across distant viewpoints and temporal instances is essential. We present a self-supervised framework to adapt a generic person appearance descriptor to the unlabeled videos by exploiting motion tracking, mutual exclusion constraints, and multi-view geometry. The adapted discriminative descriptor is used in a tracking-by-clustering formulation. We validate the effectiveness of our descriptor learning on WILDTRACK T. Chavdarova et al., "WILDTRACK: A multi-camera HD dataset for dense unscripted pedestrian detection," in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., 2018, pp. 5030-5039. and three new complex social scenes captured by multiple cameras with up to 60 people "in the wild". We report significant improvement in association accuracy (up to 18 percent) and stable and coherent 3D human skeleton tracking (5 to 10 times) over the baseline. Using the reconstructed 3D skeletons, we cut the input videos into a multi-angle video where the image of a specified person is shown from the best visible front-facing camera. Our algorithm detects inter-human occlusion to determine the camera switching moment while still maintaining the flow of the action well. Website: http://www.cs.cmu.edu/~ILIM/projects/IM/Association4Tracking.

We introduce a new framework for the exact point-wise ℓp robustness verification problem that exploits the layer-wise geometric structure of deep feed-forward networks with rectified linear activations (ReLU networks). The activation regions of the network partition the input space, and one can verify the ℓp robustness around a point by checking all the activation regions within the desired radius. The GeoCert algorithm (Jordan et al., NeurIPS 2019) treats this partition as a generic polyhedral complex in order to detect which region to check next. In contrast, our LayerCert framework considers the nested hyperplane arrangement structure induced by the layers of the ReLU network and explores regions in a hierarchical manner. We show that, under certain conditions on the algorithm parameters, LayerCert provably reduces the number and size of the convex programs that one needs to solve compared to GeoCert. Furthermore, our LayerCert framework allows the incorporation of lower bounding routines based on convex relaxations to further improve performance. Experimental results demonstrate that LayerCert can significantly reduce both the number of convex programs solved and the running time over the state-of-the-art.

In this paper, we propose a unified panoptic segmentation network (UPSNet) for tackling the newly proposed panoptic segmentation task. On top of a single backbone residual network, we first design a deformable convolution based semantic segmentation head and a Mask R-CNN style instance segmentation head which solve these two subtasks simultaneously. More importantly, we introduce a parameter-free panoptic head which solves the panoptic segmentation via pixel-wise classification. It first leverages the logits from the previous two heads and then innovatively expands the representation for enabling prediction of an extra unknown class which helps better resolving the conflicts between semantic and instance segmentation. Besides, it handles the challenge caused by the varying number of instances and permits back propagation to the bottom modules in an end-to-end manner. Extensive experimental results on Cityscapes, COCO and our internal dataset demonstrate that our UPSNet achieves state-of-the-art performance with much faster inference. Code has been made available at: https://github.com/uber-research/UPSNet.

The motion planners used in self-driving vehicles need to generate trajectories that are safe, comfortable, and obey the traffic rules. This is usually achieved by two modules: behavior planner, which handles high-level decisions and produces a coarse trajectory, and trajectory planner that generates a smooth, feasible trajectory for the duration of the planning horizon. These planners, however, are typically developed separately, and changes in the behavior planner might affect the trajectory planner in unexpected ways. Furthermore, the final trajectory outputted by the trajectory planner might differ significantly from the one generated by the behavior planner, as they do not share the same objective. In this paper, we propose a jointly learnable behavior and trajectory planner. Unlike most existing learnable motion planners that address either only behavior planning, or use an uninterpretable neural network to represent the entire logic from sensors to driving commands, our approach features an interpretable cost function on top of perception, prediction and vehicle dynamics, and a joint learning algorithm that learns a shared cost function employed by our behavior and trajectory components. Experiments on real-world self-driving data demonstrate that jointly learned planner performs significantly better in terms of both similarity to human driving and other safety metrics, compared to baselines that do not adopt joint behavior and trajectory learning.

Edges, boundaries and contours are important subjects of study in both computer graphics and computer vision. On one hand, they are the 2D elements that convey 3D shapes, on the other hand, they are indicative of occlusion events and thus separation of objects or semantic concepts. In this paper, we aim to generate contour drawings, boundary-like drawings that capture the outline of the visual scene. Prior art often cast this problem as boundary detection. However, the set of visual cues presented in the boundary detection output are different from the ones in contour drawings, and also the artistic style is ignored. We address these issues by collecting a new dataset of contour drawings and proposing a learning-based method that resolves diversity in the annotation and, unlike boundary detectors, can work with imperfect alignment of the annotation and the actual ground truth. Our method surpasses previous methods quantitatively and qualitatively. Surprisingly, when our model fine-tunes on BSDS500, we achieve the state-of-the-art performance in salient boundary detection, suggesting contour drawing might be a scalable alternative to boundary annotation, which at the same time is easier and more interesting for annotators to draw.

Abstract Lighting is a critical element of theater. A lighting designer is responsible for drawing the audience's attention to a specific part of the stage, setting time of day, creating a mood, and conveying emotions. Designers often begin the lighting design process by collecting reference visual imagery that captures different aspects of their artistic intent. Then, they experiment with various lighting options to determine which ideas work best on stage. However, modern stages contain tens to hundreds of lights, and setting each light source's parameters individually to realize an idea is both tedious and requires expert skill. In this paper, we describe an exploratory lighting design tool based on feedback from professional designers. The system extracts abstract visual objectives from reference imagery and applies them to target regions of the stage. Our system can rapidly generate plausible design candidates that embody the visual objectives through a Gibbs sampling method, and present them as a design gallery for rapid exploration and iterative refinement. We demonstrate that the resulting system allows lighting designers of all skill levels to quickly create and communicate complex designs, even for scenes containing many color‐changing lights.

Abstract Creating dynamic virtual environments consisting of humans interacting with objects is a fundamental problem in computer graphics. While it is well‐accepted that agent interactions play an essential role in synthesizing such scenes, most extant techniques exclusively focus on static scenes, leaving the dynamic component out. In this paper, we present a generative model to synthesize plausible multi‐step dynamic human‐object interactions. Generating multi‐step interactions is challenging since the space of such interactions is exponential in the number of objects, activities, and time steps. We propose to handle this combinatorial complexity by learning a lower dimensional space of plausible human‐object interactions. We use action plots to represent interactions as a sequence of discrete actions along with the participating objects and their states. To build action plots, we present an automatic method that uses state‐of‐the‐art computer vision techniques on RGB videos in order to detect individual objects and their states, extract the involved hands, and recognize the actions performed. The action plots are built from observing videos of everyday activities and are used to train a generative model based on a Recurrent Neural Network (RNN). The network learns the causal dependencies and constraints between individual actions and can be used to generate novel and diverse multi‐step human‐object interactions. Our representation and generative model allows new capabilities in a variety of applications such as interaction prediction, animation synthesis, and motion planning for a real robotic agent.

Next generation smart and augmented reality systems demand a computational understanding of monocular footage that captures humans in physical spaces to reveal plausible object arrangements and human-object interactions. Despite recent advances, both in scene layout and human motion analysis, the above setting remains challenging to analyze due to regular occlusions that occur between objects and human motions. We observe that the interaction between object arrangements and human actions is often strongly correlated, and hence can be used to help recover from these occlusions. We present iMapper, a data-driven method to identify such human-object interactions and utilize them to infer layouts of occluded objects. Starting from a monocular video with detected 2D human joint positions that are potentially noisy and occluded, we first introduce the notion of interaction-saliency as space-time snapshots where informative human-object interactions happen. Then, we propose a global optimization to retrieve and fit interactions from a database to the detected salient interactions in order to best explain the input video. We extensively evaluate the approach, both quantitatively against manually annotated ground truth and through a user study, and demonstrate that iMapper produces plausible scene layouts for scenes with medium to heavy occlusion. Code and data are available on the project page.

Discovering 3D arrangements of objects from single indoor images is important given its many applications such as interior design and content creation for virtual environments. Although heavily researched in the recent years, existing approaches break down under medium to heavy occlusion as the core image-space region detection module fails in absence of directly visible cues. Instead, we take into account holistic contextual 3D information, exploiting the fact that objects in indoor scenes co-occur mostly in typical configurations. First, we use a neural network trained on real indoor annotated images to extract 2D keypoints, and feed them to a 3D candidate object generation stage. Then, we solve a global selection problem among these candidates using pairwise co-occurrence statistics discovered from a large 3D scene database. We iterate the process allowing for candidates with low keypoint response to be incrementally detected based on the location of the already discovered nearby objects. We demonstrate significant performance improvement over combinations of state-of-the-art methods, especially for scenes with moderately to severely occluded objects.

Video summaries are a popular way to share important events, but creating good summaries is hard. It requires expertise in both capturing and editing footage. While hiring a professional videographer is possible, this is too costly for most casual events. An alternative is to place 360 video cameras around an event space to capture footage passively and then extract regular field-of-view (RFOV) shots for the summary. This paper focuses on the problem of extracting such RFOV shots. Since we cannot actively control the cameras or the scene, it is hard to create "ideal' shots that adhere strictly to traditional cinematography rules. To better understand the tradeoffs, we study human preferences for static and moving camera RFOV shots generated from 360 footage. From the findings, we derive design guidelines. As a secondary contribution, we use these guidelines to develop automatic algorithms that we demonstrate in a prototype user interface for extracting RFOV shots from 360 videos.

We address the problem of finding realistic geometric corrections to a foreground object such that it appears natural when composited into a background image. To achieve this, we propose a novel Generative Adversarial Network (GAN) architecture that utilizes Spatial Transformer Networks (STNs) as the generator, which we call Spatial Transformer GANs (ST-GANs). ST-GANs seek image realism by operating in the geometric warp parameter space. In particular, we exploit an iterative STN warping scheme and propose a sequential training strategy that achieves better results compared to naive training of a single generator. One of the key advantages of ST-GAN is its applicability to high-resolution images indirectly since the predicted warp parameters are transferable between reference frames. We demonstrate our approach in two applications: (1) visualizing how indoor furniture (e.g. from product images) might be perceived in a room, (2) hallucinating how accessories like glasses would look when matched with real portraits.

This paper proposes a deep neural network (DNN) for piece-wise planar depthmap reconstruction from a single RGB image. While DNNs have brought remarkable progress to single-image depth prediction, piece-wise planar depthmap reconstruction requires a structured geometry representation, and has been a difficult task to master even for DNNs. The proposed end-to-end DNN learns to directly infer a set of plane parameters and corresponding plane segmentation masks from a single RGB image. We have generated more than 50,000 piece-wise planar depthmaps for training and testing from ScanNet, a large-scale RGBD video database. Our qualitative and quantitative evaluations demonstrate that the proposed approach outperforms baseline methods in terms of both plane segmentation and depth estimation accuracy. To the best of our knowledge, this paper presents the first end-to-end neural architecture for piece-wise planar reconstruction from a single RGB image. Code and data are available at https://github.com/art-programmer/PlaneNet.

Recently, researchers have developed black-box approaches to mine design and interaction data from mobile apps. Although the data captured during this interaction mining is descriptive, it does not expose the design semantics of UIs: what elements on the screen mean and how they are used. This paper introduces an automatic approach for generating semantic annotations for mobile app UIs. Through an iterative open coding of 73k UI elements and 720 screens, we contribute a lexical database of 25 types of UI components, 197 text button concepts, and 135 icon classes shared across apps. We use this labeled data to learn code-based patterns to detect UI components and to train a convolutional neural network that distinguishes between icon classes with 94% accuracy. To demonstrate the efficacy of our approach at scale, we compute semantic annotations for the 72k unique UIs in the Rico dataset, assigning labels for 78% of the total visible, non-redundant elements.

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.

We present a transformation-grounded image generation network for novel 3D view synthesis from a single image. Our approach first explicitly infers the parts of the geometry visible both in the input and novel views and then casts the remaining synthesis problem as image completion. Specifically, we both predict a flow to move the pixels from the input to the novel view along with a novel visibility map that helps deal with occulsion/disocculsion. Next, conditioned on those intermediate results, we hallucinate (infer) parts of the object invisible in the input image. In addition to the new network structure, training with a combination of adversarial and perceptual loss results in a reduction in common artifacts of novel view synthesis such as distortions and holes, while successfully generating high frequency details and preserving visual aspects of the input image. We evaluate our approach on a wide range of synthetic and real examples. Both qualitative and quantitative results show our method achieves significantly better results compared to existing methods.

Traditional face editing methods often require a number of sophisticated and task specific algorithms to be applied one after the other - a process that is tedious, fragile, and computationally intensive. In this paper, we propose an end-to-end generative adversarial network that infers a face-specific disentangled representation of intrinsic face properties, including shape (i.e. normals), albedo, and lighting, and an alpha matte. We show that this network can be trained on “in-the-wild” images by incorporating an in-network physically-based image formation module and appropriate loss functions. Our disentangling latent representation allows for semantically relevant edits, where one aspect offacial appearance can be manipulated while keeping orthogonal properties fixed, and we demonstrate its use for a number offacial editing applications.

We present an end-to-end, multimodal, fully convolutional network for extracting semantic structures from document images. We consider document semantic structure extraction as a pixel-wise segmentation task, and propose a unified model that classifies pixels based not only on their visual appearance, as in the traditional page segmentation task, but also on the content of underlying text. Moreover, we propose an efficient synthetic document generation process that we use to generate pretraining data for our network. Once the network is trained on a large set of synthetic documents, we fine-tune the network on unlabeled real documents using a semi-supervised approach. We systematically study the optimum network architecture and show that both our multimodal approach and the synthetic data pretraining significantly boost the performance.

Indoor scene understanding is central to applications such as robot navigation and human companion assistance. Over the last years, data-driven deep neural networks have outperformed many traditional approaches thanks to their representation learning capabilities. One of the bottlenecks in training for better representations is the amount of available per-pixel ground truth data that is required for core scene understanding tasks such as semantic segmentation, normal prediction, and object boundary detection. To address this problem, a number of works proposed using synthetic data. However, a systematic study of how such synthetic data is generated is missing. In this work, we introduce a large-scale synthetic dataset with 500K physically-based rendered images from 45K realistic 3D indoor scenes. We study the effects of rendering methods and scene lighting on training for three computer vision tasks: surface normal prediction, semantic segmentation, and object boundary detection. This study provides insights into the best practices for training with synthetic data (more realistic rendering is worth it) and shows that pretraining with our new synthetic dataset can improve results beyond the current state of the art on all three tasks.

The ability to edit materials of objects in images is desirable by many content creators. However, this is an extremely challenging task as it requires to disentangle intrinsic physical properties of an image. We propose an end-to-end network architecture that replicates the forward image formation process to accomplish this task. Specifically, given a single image, the network first predicts intrinsic properties, i.e. shape, illumination, and material, which are then provided to a rendering layer. This layer performs in-network image synthesis, thereby enabling the network to understand the physics behind the image formation process. The proposed rendering layer is fully differentiable, supports both diffuse and specular materials, and thus can be applicable in a variety of problem settings. We demonstrate a rich set of visually plausible material editing examples and provide an extensive comparative study.

The success of various applications including robotics, digital content creation, and visualization demand a structured and abstract representation of the 3D world from limited sensor data. Inspired by the nature of human perception of 3D shapes as a collection of simple parts, we explore such an abstract shape representation based on primitives. Given a single depth image of an object, we present 3DPRNN, a generative recurrent neural network that synthesizes multiple plausible shapes composed of a set of primitives. Our generative model encodes symmetry characteristics of common man-made objects, preserves long-range structural coherence, and describes objects of varying complexity with a compact representation. We also propose a method based on Gaussian Fields to generate a large scale dataset of primitive-based shape representations to train our network. We evaluate our approach on a wide range of examples and show that it outperforms nearest-neighbor based shape retrieval methods and is on-par with voxelbased generative models while using a significantly reduced parameter space.

Current state-of-the-art solutions for motion capture from a single camera are optimization driven: they optimize the parameters of a 3D human model so that its re-projection matches measurements in the video (e.g. person segmentation, optical flow, keypoint detections etc.). Optimization models are susceptible to local minima. This has been the bottleneck that forced using clean green-screen like backgrounds at capture time, manual initialization, or switching to multiple cameras as input resource. In this work, we propose a learning based motion capture model for single camera input. Instead of optimizing mesh and skeleton parameters directly, our model optimizes neural network weights that predict 3D shape and skeleton configurations given a monocular RGB video. Our model is trained using a combination of strong supervision from synthetic data, and self-supervision from differentiable rendering of (a) skeletal keypoints, (b) dense 3D mesh motion, and (c) human-background segmentation, in an end-to-end framework. Empirically we show our model combines the best of both worlds of supervised learning and test-time optimization: supervised learning initializes the model parameters in the right regime, ensuring good pose and surface initialization at test time, without manual effort. Self-supervision by back-propagating through differentiable rendering allows (unsupervised) adaptation of the model to the test data, and offers much tighter fit than a pretrained fixed model. We show that the proposed model improves with experience and converges to low-error solutions where previous optimization methods fail.

We propose an automatic method to infer high dynamic range illumination from a single, limited field-of-view, low dynamic range photograph of an indoor scene. In contrast to previous work that relies on specialized image capture, user input, and/or simple scene models, we train an end-to-end deep neural network that directly regresses a limited field-of-view photo to HDR illumination, without strong assumptions on scene geometry, material properties, or lighting. We show that this can be accomplished in a three step process: 1) we train a robust lighting classifier to automatically annotate the location of light sources in a large dataset of LDR environment maps, 2) we use these annotations to train a deep neural network that predicts the location of lights in a scene from a single limited field-of-view photo, and 3) we fine-tune this network using a small dataset of HDR environment maps to predict light intensities. This allows us to automatically recover high-quality HDR illumination estimates that significantly outperform previous state-of-the-art methods. Consequently, using our illumination estimates for applications like 3D object insertion, produces photo-realistic results that we validate via a perceptual user study.

We introduce Skippy, a novel algorithm for 3D interactive curve modeling from a single view. While positing curves in space can be a tedious task, our rapid sketching algorithm allows users to draw curves in and around existing geometry in a controllable manner. The key insight behind our system is to automatically infer the 3D curve coordinates by enumerating a large set of potential curve trajectories. More specifically, we partition 2D strokes into continuous segments that land both on and off the geometry, duplicating segments that could be placed in front or behind, to form a directed graph. We use distance fields to estimate 3D coordinates for our curve segments and solve for an optimally smooth path that follows the curvature of the scene geometry while avoiding intersections. Using our curve design framework we present a collection of novel editing operations allowing artists to rapidly explore and refine the combinatorial space of solutions. Furthermore, we include the quick placement of transient geometry to aid in guiding the 3D curve. Finally we demonstrate our interactive design curve system on a variety of applications including geometric modeling, and camera motion path planning.

We introduce a novel neural network architecture for encoding and synthesis of 3D shapes, particularly their structures. Our key insight is that 3D shapes are effectively characterized by their hierarchical organization of parts, which reflects fundamental intra-shape relationships such as adjacency and symmetry. We develop a recursive neural net (RvNN) based autoencoder to map a flat, unlabeled, arbitrary part layout to a compact code. The code effectively captures hierarchical structures of man-made 3D objects of varying structural complexities despite being fixed-dimensional: an associated decoder maps a code back to a full hierarchy. The learned bidirectional mapping is further tuned using an adversarial setup to yield a generative model of plausible structures, from which novel structures can be sampled. Finally, our structure synthesis framework is augmented by a second trained module that produces fine-grained part geometry, conditioned on global and local structural context, leading to a full generative pipeline for 3D shapes. We demonstrate that without supervision, our network learns meaningful structural hierarchies adhering to perceptual grouping principles, produces compact codes which enable applications such as shape classification and partial matching, and supports shape synthesis and interpolation with significant variations in topology and geometry.

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.

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.

Procedural modeling techniques can produce high quality visual content through complex rule sets. However, controlling the outputs of these techniques for design purposes is often notoriously difficult for users due to the large number of parameters involved in these rule sets and also their non-linear relationship to the resulting content. To circumvent this problem, we present a sketch-based approach to procedural modeling. Given an approximate and abstract hand-drawn 2D sketch provided by a user, our algorithm automatically computes a set of procedural model parameters, which in turn yield multiple, detailed output shapes that resemble the user's input sketch. The user can then select an output shape, or further modify the sketch to explore alternative ones. At the heart of our approach is a deep Convolutional Neural Network (CNN) that is trained to map sketches to procedural model parameters. The network is trained by large amounts of automatically generated synthetic line drawings. By using an intuitive medium, i.e., freehand sketching as input, users are set free from manually adjusting procedural model parameters, yet they are still able to create high quality content. We demonstrate the accuracy and efficacy of our method in a variety of procedural modeling scenarios including design of man-made and organic shapes.

Understanding the 3D world is a fundamental problem in computer vision. However, learning a good representation of 3D objects is still an open problem due to the high dimensionality of the data and many factors of variation involved. In this work, we investigate the task of single-view 3D object reconstruction from a learning agent's perspective. We formulate the learning process as an interaction between 3D and 2D representations and propose an encoder-decoder network with a novel projection loss defined by the projective transformation. More importantly, the projection loss enables the unsupervised learning using 2D observation without explicit 3D supervision. We demonstrate the ability of the model in generating 3D volume from a single 2D image with three sets of experiments: (1) learning from single-class objects; (2) learning from multi-class objects and (3) testing on novel object classes. Results show superior performance and better generalization ability for 3D object reconstruction when the projection loss is involved.

Human motion is complex and difficult to synthesize realistically. Automatic style transfer to transform the mood or identity of a character's motion is a key technology for increasing the value of already synthesized or captured motion data. Typically, state-of-the-art methods require all independent actions observed in the input to be present in a given style database to perform realistic style transfer. We introduce a spectral style transfer method for human motion between independent actions, thereby greatly reducing the required effort and cost of creating such databases. We leverage a spectral domain representation of the human motion to formulate a spatial correspondence free approach. We extract spectral intensity representations of reference and source styles for an arbitrary action, and transfer their difference to a novel motion which may contain previously unseen actions. Building on this core method, we introduce a temporally sliding window filter to perform the same analysis locally in time for heterogeneous motion processing. This immediately allows our approach to serve as a style database enhancement technique to fill-in non-existent actions in order to increase previous style transfer method's performance. We evaluate our method both via quantitative experiments, and through administering controlled user studies with respect to previous work, where significant improvement is observed with our approach.

Procedural modeling systems allow users to create high quality content through parametric, conditional or stochastic rule sets. While such approaches create an abstraction layer by freeing the user from direct geometry editing, the nonlinear nature and the high number of parameters associated with such design spaces result in arduous modeling experiences for non-expert users. We propose a method to enable intuitive exploration of such high dimensional procedural modeling spaces within a lower dimensional space learned through autoencoder network training. Our method automatically generates a representative training dataset from the procedural modeling rule set based on shape similarity features. We then leverage the samples in this dataset to train an autoencoder neural network, while also structuring the learned lower dimensional space for continuous exploration with respect to shape features. We demonstrate the efficacy our method with user studies where designers create content with more than 10-fold faster speeds using our system compared to the classic procedural modeling interface.

We propose a shape editing method where the user creates geometric deformations using a set of semantic attributes, thus avoiding the need for detailed geometric manipulations. In contrast to prior work, we focus on continuous deformations instead of discrete part substitutions. Our method provides a platform for quick design explorations and allows non-experts to produce semantically guided shape variations that are otherwise difficult to attain. We crowdsource a large set of pairwise comparisons between the semantic attributes and geometry and use this data to learn a continuous mapping from the semantic attributes to geometry. The resulting map enables simple and intuitive shape manipulations based solely on the learned attributes. We demonstrate our method on large datasets using two different user interaction modes and evaluate its usability with a set of user studies.

For large collections of 3D models, the ability to detect and localize parts of interest is necessary to provide search and visualization enhancements beyond simple high-level categorization. While current 3D labeling approaches rely on learning from fully labeled meshes, such training data is difficult to acquire at scale. In this work we explore learning to detect object parts from sparsely labeled data, i.e. we operate under the assumption that for any object part we have only one labeled vertex rather than a full region segmentation. Similarly, we also learn to output a single representative vertex for each detected part. Such localized predictions are useful for applications where visualization is important. Our approach relies heavily on exploiting the spatial configuration of parts on a model to drive the detection. Inspired by structured multi-class object detection models for images, we develop an algorithm that combines independently trained part classifiers with a structured SVM model, and show promising results on real-world textured 3D data.

We present a novel co-segmentation method for textured 3D shapes. Our algorithm takes a collection of textured shapes belonging to the same category and sparse annotations of foreground segments, and produces a joint dense segmentation of the shapes in the collection. We model the segments by a collectively trained Gaussian mixture model. The final model segmentation is formulated as an energy minimization across all models jointly, where intra-model edges control the smoothness and separation of model segments, and inter-model edges impart global consistency. We show promising results on two large real-world datasets, and also compare with previous shape-only 3D segmentation methods using publicly available datasets.

We present a method for learning custom deformation handles for an object, from a co-analysis of similar objects. Our approach identifies the geometric and spatial constraints among the different parts of an object, and makes this information available through abstract shape handles. These handles allow the user to prescribe arbitrary deformation directives including free-form surface deformations. However, only a subset of admissible deformations is enabled to the user as learned from the constraint space. Example applications are presented in shape editing, co-deformation and style transfer.

We present a co-abstraction method that takes as input a collection of 3D objects, and produces a mutually consistent and individually identity-preserving abstraction of each object. In general, an abstraction is a simpler version of a shape that preserves its main characteristics. We hypothesize, however, that there is no single abstraction of an object. Instead, there is a variety of possible abstractions, and an admissible one can only be chosen conjointly with other objects' abstractions. To this end, we introduce a new approach that hierarchically generates a spectrum of abstractions for each model in a shape collection. Given the spectra, we compute the appropriate abstraction level for each model such that shape simplification and inter-set consistency are collectively maximized, while individual shape identities are preserved.