Pattern images are everywhere in the digital and physical worlds, and tools to edit them are valuable. But editing pattern images is tricky: desired edits are often programmatic: structure-aware edits that alter the underlying program which generates the pattern. One could attempt to infer this underlying program, but current methods for doing so struggle with complex images and produce unorganized programs that make editing tedious. In this work, we introduce a novel approach to perform programmatic edits on pattern images. By using a pattern analogy—a pair of simple patterns to demonstrate the intended edit—and a learning-based generative model to execute these edits, our method allows users to intuitively edit patterns. To enable this paradigm, we introduce SplitWeave, a domain-specific language that, combined with a framework for sampling synthetic pattern analogies, enables the creation of a large, high-quality synthetic training dataset. We also present TriFuser, a Latent Diffusion Model (LDM) designed to overcome critical issues that arise when naively deploying LDMs to this task. Extensive experiments on real-world, artist-sourced patterns reveals that our method faithfully performs the demonstrated edit while also generalizing to related pattern styles beyond its training distribution.

Current controls over diffusion models (e.g., through text or ControlNet) for image generation fall short in recognizing abstract, continuous attributes like illumination direction or non-rigid shape change. In this paper, we present an approach for allowing users of text-to-image models to have fine-grained control of several attributes in an image. We do this by engineering special sets of input tokens that can be transformed in a continuous manner – we call them Continuous 3D Words. These attributes can, for example, be represented as sliders and applied jointly with text prompts for fine-grained control over image generation. Given only a single mesh and a rendering engine, we show that our approach can be adopted to provide continuous user control over several 3D-aware attributes, including time-of-day illumination, bird wing orientation, dollyzoom effect, and object poses. Our method is capable of conditioning image creation with multiple Continuous 3D Words and text descriptions simultaneously while adding no overhead to the generative process. Project Page: https://ttchengab.github.io/continuous_3d_words

We present ANISE, a method that reconstructs a 3D shape from partial observations (images or sparse point clouds) using a part-aware neural implicit shape representation. The shape is formulated as an assembly of neural implicit functions, each representing a different part instance. In contrast to previous approaches, the prediction of this representation proceeds in a coarse-to-fine manner. Our model first reconstructs a structural arrangement of the shape in the form of geometric transformations of its part instances. Conditioned on them, the model predicts part latent codes encoding their surface geometry. Reconstructions can be obtained in two ways: (i) by directly decoding the part latent codes to part implicit functions, then combining them into the final shape; or (ii) by using part latents to retrieve similar part instances in a part database and assembling them in a single shape. We demonstrate that, when performing reconstruction by decoding part representations into implicit functions, our method achieves state-of-the-art part-aware reconstruction results from both images and sparse point clouds. When reconstructing shapes by assembling parts retrieved from a dataset, our approach significantly outperforms traditional shape retrieval methods even when significantly restricting the database size. We present our results in well-known sparse point cloud reconstruction and single-view reconstruction benchmarks.

We present 3DMiner – a pipeline for mining 3D shapes from challenging large-scale unannotated image datasets. Unlike other unsupervised 3D reconstruction methods, we assume that, within a large-enough dataset, there must exist images of objects with similar shapes but varying backgrounds, textures, and viewpoints. Our approach leverages the recent advances in learning self-supervised image representations to cluster images with geometrically similar shapes and find common image correspondences between them. We then exploit these correspondences to obtain rough camera estimates as initialization for bundle-adjustment. Finally, for every image cluster, we apply a progressive bundle-adjusting reconstruction method to learn a neural occupancy field representing the underlying shape. We show that this procedure is robust to several types of errors introduced in previous steps (e.g., wrong camera poses, images containing dissimilar shapes, etc.), allowing us to obtain shape and pose annotations for images in-the-wild. When using images from Pix3D chairs, our method is capable of producing significantly better results than state-of-the-art unsupervised 3D reconstruction techniques, both quantitatively and qualitatively. Furthermore, we show how 3DMiner can be applied to in-the-wild data by reconstructing shapes present in images from the LAION-5B dataset. Project Page: https://ttchengab.github.io/3dminerOfficial.

Placing and orienting a camera to compose aesthetically meaningful shots of a scene is not only a key objective in real-world photography and cinematography but also for virtual content creation. The framing of a camera often significantly contributes to the story telling in movies, games, and mixed reality applications. Generating single camera poses or even contiguous trajectories either requires a significant amount of manual labor or requires solving high-dimensional optimization problems, which can be computationally demanding and error-prone. In this paper, we introduce GAIT, a framework for training a Deep Reinforcement Learning (DRL) agent, that learns to automatically control a camera to generate a sequence of aesthetically meaningful views for synthetic 3D indoor scenes. To generate sequences of frames with high aesthetic value, GAIT relies on a neural aesthetics estimator, which is trained on a crowed-sourced dataset. Additionally, we introduce regularization techniques for diversity and smoothness to generate visually interesting trajectories for a 3D environment, and to constrain agent acceleration in the reward function to generate a smooth sequence of camera frames. We validated our method by comparing it to baseline algorithms, based on a perceptual user study, and through ablation studies. Code and visual results are available on the project website: https://desaixie.github.io/gait-rl

Three-dimensional curve drawing in Augmented Reality (AR) enables users to create 3D curves that fit within the real-world scene. It has applications in 3D design, sculpting, and animation. However, the task complexity increases when the desirable path for the curve is obstructed by the physical environment or by what the camera can see. For example, it is difficult to draw a curve that wraps around an object or scales to out-of-reach places. We propose WARPY, an environment-aware 3D curve drawing tool for mobile AR. Our system enables users to draw freeform curves from a distance in AR by combining 2D-to-3D sketch inference with geometric proxies. Geometric Proxies can be obtained via 3D scanning or from a list of pre-defined primitives. WARPY also provides a multi-view mode to enable users to sketch a curve from multiple viewpoints, which is useful if the target curve cannot fit within the camera's field of view. We conducted two user studies and found that WARPY can be a viable tool to help users create complex and large curves in AR.

Abstract Procedural modeling allows for an automatic generation of large amounts of similar assets, but there is limited control over the generated output. We address this problem by introducing Automatic Differentiable Procedural Modeling (ADPM). The forward procedural model generates a final editable model. The user modifies the output interactively, and the modifications are transferred back to the procedural model as its parameters by solving an inverse procedural modeling problem. We present an auto‐differentiable representation of the procedural model that significantly accelerates optimization. In ADPM the procedural model is always available, all changes are non‐destructive, and the user can interactively model the 3D object while keeping the procedural representation. ADPM provides the user with precise control over the resulting model comparable to non‐procedural interactive modeling. ADPM is node‐based, and it generates hierarchical 3D scene geometry converted to a differentiable computational graph. Our formulation focuses on the differentiability of high‐level primitives and bounding volumes of components of the procedural model rather than the detailed mesh geometry. Although this high‐level formulation limits the expressiveness of user edits, it allows for efficient derivative computation and enables interactivity. We designed a new optimizer to solve for inverse procedural modeling. It can detect that an edit is under‐determined and has degrees of freedom. Leveraging cheap derivative evaluation, it can explore the region of optimality of edits and suggest various configurations, all of which achieve the requested edit differently. We show our system's efficiency on several examples, and we validate it by a user study.

Procedural material graphs are a compact, parameteric, and resolution-independent representation that are a popular choice for material authoring. However, designing procedural materials requires significant expertise and publicly accessible libraries contain only a few thousand such graphs. We present MatFormer, a generative model that can produce a diverse set of high-quality procedural materials with complex spatial patterns and appearance. While procedural materials can be modeled as directed (operation) graphs, they contain arbitrary numbers of heterogeneous nodes with unstructured, often long-range node connections, and functional constraints on node parameters and connections. MatFormer addresses these challenges with a multi-stage transformer-based model that sequentially generates nodes, node parameters, and edges, while ensuring the semantic validity of the graph. In addition to generation, MatFormer can be used for the auto-completion and exploration of partial material graphs. We qualitatively and quantitatively demonstrate that our method outperforms alternative approaches, in both generated graph and material quality.

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.

We propose DeepMetaHandles, a 3D conditional generative model based on mesh deformation. Given a collection of 3D meshes of a category and their deformation handles (control points), our method learns a set of meta-handles for each shape, which are represented as combinations of the given handles. The disentangled meta-handles factorize all the plausible deformations of the shape, while each of them corresponds to an intuitive deformation. A new deformation can then be generated by sampling the co-efficients of the meta-handles in a specific range. We employ biharmonic coordinates as the deformation function, which can smoothly propagate the control points’ translations to the entire mesh. To avoid learning zero deformaion as meta-handles, we incorporate a target-fitting module which deforms the input mesh to match a random target. To enhance deformations’ plausibility, we employ a soft-rasterizer-based discriminator that projects the meshes to a 2D space. Our experiments demonstrate the superiority of the generated deformations as well as the interpretability and consistency of the learned meta-handles. The code is available at https://github.com/Colin97/DeepMetaHandles.

Representing human-made objects as a collection of base primitives has a long history in computer vision and reverse engineering. In the case of high-resolution point cloud scans, the challenge is to be able to detect both large primitives as well as those explaining the detailed parts. While the classical RANSAC approach requires case-specific parameter tuning, state-of-the-art networks are limited by memory consumption of their backbone modules such as PointNet++ [27], and hence fail to detect the fine-scale primitives. We present Cascaded Primitive Fitting Networks (CPFN) that relies on an adaptive patch sampling network to assemble detection results of global and local primitive detection networks. As a key enabler, we present a merging formulation that dynamically aggregates the primitives across global and local scales. Our evaluation demonstrates that CPFN improves the state-of-the-art SPFN performance by 13 − 14% on high-resolution point cloud datasets and specifically improves the detection of fine-scale primitives by 20 − 22%. Our code is available at: https://github.com/erictuanle/CPFN

Abstract We review recent methods in 2D creative pattern generation and their control mechanisms, focusing on procedural methods. The review is motivated by an artist's perspective and investigates interactive pattern generation as a complex design problem. While the repetitive nature of patterns is well‐suited to algorithmic creation and automation, an artist needs more flexible control mechanisms for adaptable and inventive designs. We organize the state of the art around pattern design features, such as repetition, frames, curves, directionality, and single visual accents. Within those areas, we summarize and discuss the techniques' control mechanisms for enabling artist intent. The discussion includes questions of how input is given by the artist, what type of content the artist inputs, where the input affects the canvas spatially, and when input can be given in the timeline of the creation process. We categorize the available control mechanisms on an algorithmic level and categorize their input modes based on exemplars, parameterization, handling, filling, guiding, and placing interactions. To better understand the potential of the current techniques for creative design and to make such an investigation more manageable, we motivate our discussion with how navigation, transparency, variation, and stimulation enable creativity. We conclude our review by identifying possible new directions that can inspire innovation for artist‐centered creation processes and algorithms.

Detecting segments of interest from videos is a common problem for many applications. And yet it is a challenging problem as it often requires not only knowledge of individual target segments, but also contextual understanding of the entire video and the relationships between the target segments. To address this problem, we propose the Sequence-to-Segments Network (S2N), a novel and general end-to-end sequential encoder-decoder architecture. S2N first encodes the input video into a sequence of hidden states that capture information progressively, as it appears in the video. It then employs the Segment Detection Unit (SDU), a novel decoding architecture, that sequentially detects segments. At each decoding step, the SDU integrates the decoder state and encoder hidden states to detect a target segment. During training, we address the problem of finding the best assignment of predicted segments to ground truth using the Hungarian Matching Algorithm with Lexicographic Cost. Additionally we propose to use the squared Earth Mover's Distance to optimize the localization errors of the segments. We show the state-of-the-art performance of S2N across numerous tasks, including video highlighting, video summarization, and human action proposal generation.

Procedural modeling has produced amazing results, yet fundamental issues such as controllability and limited user guidance persist. We introduce a novel procedural model called PICO (Procedural Iterative Constrained Optimizer) and PICO-Graph that is the underlying procedural model designed with optimization in mind. The key novelty of PICO is that it enables the exploration of generative designs by combining both user and environmental constraints into a single framework by using optimization without the need to write procedural rules. The PICO-Graph procedural model consists of a set of geometry generating operations and a set of axioms connected in a directed cyclic graph. The forward generation is initiated by a set of axioms that use the connections to send coordinate systems and geometric objects through the PICO-Graph, which in turn generates more objects. This allows for fast generation of complex and varied geometries. Moreover, we combine PICO-Graph with efficient optimization that allows for quick exploration of the generated models and the generation of variants. The user defines the rules, the axioms, and the set of constraints; for example, whether an existing object should be supported by the generated model, whether symmetries exist, whether the object should spin, etc. PICO then generates a class of geometric models and optimizes them so that they fulfill the constraints. The generation and the optimization in our implementation provides interactive user control during model execution providing continuous feedback. For example, the user can sketch the constraints and guide the geometry to meet these specified goals. We show PICO on a variety of examples such as the generation of procedural chairs with multiple supports, generation of support structures for 3D printing, generation of spinning objects, or generation of procedural terrains matching a given input. Our framework could be used as a component in a larger design workflow; its strongest application is in the early rapid ideation and prototyping phases.

Photographic composition is often taught as alignment with composition grids-most commonly, the rule of thirds. Professional photographers use more complex grids, like the harmonic armature, to achieve more diverse dynamic compositions. We are interested in understanding whether these complex grids are helpful to amateurs.

Programming offers new opportunities for visual art creation, but understanding and manipulating the abstract representations that make programming powerful can pose challenges for artists who are accustomed to manual tools and concrete visual interaction. We hypothesize that we can reduce these barriers through programming environments that link state to visual artwork output. We created Demystified Dynamic Brushes (DDB), a tool that bidirectionally links code, numerical data, and artwork across the programming interface and the execution environment - i.e., the artist's in-progress artwork. DDB automatically records stylus input as artists draw, and stores a history of brush state and output in relation to the input. This structure enables artists to inspect current and past numerical input, state, and output and control program execution through the direct selection of visual geometric elements in the drawing canvas. An observational study suggests that artists engage in program inspection when they can visually access geometric state information on the drawing canvas in the process of manual drawing.

We present a generative model to synthesize 3D shapes as sets of handles -- lightweight proxies that approximate the original 3D shape -- for applications in interactive editing, shape parsing, and building compact 3D representations. Our model can generate handle sets with varying cardinality and different types of handles. Key to our approach is a deep architecture that predicts both the parameters and existence of shape handles and a novel similarity measure that can easily accommodate different types of handles, such as cuboids or sphere-meshes. We leverage the recent advances in semantic 3D annotation as well as automatic shape summarization techniques to supervise our approach. We show that the resulting shape representations are not only intuitive, but achieve superior quality than previous state-of-the-art. Finally, we demonstrate how our method can be used in applications such as interactive shape editing and completion, leveraging the latent space learned by our model to guide these tasks.

People usually take short videos to record meaningful moments in their lives. However, selecting the most representative frame, which not only has high image visual quality but also captures video content, from a short video to share or keep is a time-consuming process for one may need to manually go through all the frames in a video to make a decision. In this paper, we introduce the problem of the best frame selection in a short video and aim to solve it automatically. Towards this end, we collect and will release a diverse large-scale short video dataset that includes 11, 000 videos shoot in our daily life. All videos are assumed to be short (e.g., a few seconds) and each video has human-annotated of the best frame. Then we introduce a deep convolutional neural network (CNN) based approach with ranking objective to automatically pick the best frame from frame sequences extracted via short videos. Additionally, we propose new evaluation metrics, especially for the best frame selection. In experiments, we show our approach outperforms various other methods significantly.

We present MATch , a method to automatically convert photographs of material samples into production-grade procedural material models. At the core of MATch is a new library DiffMat that provides differentiable building blocks for constructing procedural materials, and automatic translation of large-scale procedural models, with hundreds to thousands of node parameters, into differentiable node graphs. Combining these translated node graphs with a rendering layer yields an end-to-end differentiable pipeline that maps node graph parameters to rendered images. This facilitates the use of gradient-based optimization to estimate the parameters such that the resulting material, when rendered, matches the target image appearance, as quantified by a style transfer loss. In addition, we propose a deep neural feature-based graph selection and parameter initialization method that efficiently scales to a large number of procedural graphs. We evaluate our method on both rendered synthetic materials and real materials captured as flash photographs. We demonstrate that MATch can reconstruct more accurate, general, and complex procedural materials compared to the state-of-the-art. Moreover, by producing a procedural output, we unlock capabilities such as constructing arbitrary-resolution material maps and parametrically editing the material appearance.

Instant photo taking and sharing has become one of the most popular forms of social networking. However, taking high-quality photos is difficult as it requires knowledge and skill in photography that most non-expert users lack. In this paper we present SmartEye, a novel mobile system to help users take photos with good compositions in-situ. The back-end of SmartEye integrates the View Proposal Network (VPN), a deep learning based model that outputs composition suggestions in real time, and a novel, interactively updated module (P-Module) that adjusts the VPN outputs to account for personalized composition preferences. We also design a novel interface with functions at the front-end to enable real-time and informative interactions for photo taking. We conduct two user studies to investigate SmartEye qualitatively and quantitatively. Results show that SmartEye effectively models and predicts personalized composition preferences, provides instant high-quality compositions in-situ, and outperforms the non-personalized systems significantly.

Applications in virtual and augmented reality create a demand for rapid creation and easy access to large sets of 3D models. An effective way to address this demand is to edit or deform existing 3D models based on a reference, e.g., a 2D image which is very easy to acquire. Given such a source 3D model and a target which can be a 2D image, 3D model, or a point cloud acquired as a depth scan, we introduce 3DN, an end-to-end network that deforms the source model to resemble the target. Our method infers per-vertex offset displacements while keeping the mesh connectivity of the source model fixed. We present a training strategy which uses a novel differentiable operation, mesh sampling operator, to generalize our method across source and target models with varying mesh densities. Mesh sampling operator can be seamlessly integrated into the network to handle meshes with different topologies. Qualitative and quantitative results show that our method generates higher quality results compared to the state-of-the art learning-based methods for 3D shape generation.

Reconstructing 3D shapes from single-view images has been a long-standing research problem. In this paper, we present DISN, a Deep Implicit Surface Net- work which can generate a high-quality detail-rich 3D mesh from a 2D image by predicting the underlying signed distance fields. In addition to utilizing global image features, DISN predicts the projected location for each 3D point on the 2D image and extracts local features from the image feature maps. Combin- ing global and local features significantly improves the accuracy of the signed distance field prediction, especially for the detail-rich areas. To the best of our knowledge, DISN is the first method that constantly captures details such as holes and thin structures present in 3D shapes from single-view images. DISN achieves the state-of-the-art single-view reconstruction performance on a variety of shape categories reconstructed from both synthetic and real images. Code is available at https://github.com/laughtervv/DISN. The supplemen- tary can be found at https://xharlie.github.io/images/neurips_ 2019_supp.pdf

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.

Procedural art, or art made with programming, suggests opportunities to extend traditional arts like painting and drawing; however, this potential is limited by tools that conflict with manual practices. Programming languages present learning barriers and manual drawing input is not a first class primitive in common programming models. We hypothesize that by developing programming languages and environments that align with how manual artists work, we can build procedural systems that enhance, rather than displace, manual art. To explore this, we developed Dynamic Brushes, a programming and drawing environment motivated by interviews with artists. Dynamic Brushes enables the creation of ad-hoc drawing tools that transform stylus inputs to procedural patterns. Applications range from transforming individual strokes to behaviors that draw multiple strokes simultaneously, respond to temporal events, and leverage external data. Results from an extended evaluation with artists provide guidelines for learnable, expressive systems that blend manual and procedural creation.

Finding views with good photo composition is a challenging task for machine learning methods. A key difficulty is the lack of well annotated large scale datasets. Most existing datasets only provide a limited number of annotations for good views, while ignoring the comparative nature of view selection. In this work, we present the first large scale Comparative Photo Composition dataset, which contains over one million comparative view pairs annotated using a cost-effective crowdsourcing workflow. We show that these comparative view annotations are essential for training a robust neural network model for composition. In addition, we propose a novel knowledge transfer framework to train a fast view proposal network, which runs at 75+ FPS and achieves state-of-the-art performance in image cropping and thumbnail generation tasks on three benchmark datasets. The superiority of our method is also demonstrated in a user study on a challenging experiment, where our method significantly outperforms the baseline methods in producing diversified well-composed views.

While many approaches have been proposed to estimate and remove blur in a photo, few efforts were made to have an algorithm automatically understand the blur desirability: whether the blur is desired or not, and how it affects the quality of the photo. Such a task not only relies on low-level visual features to identify blurry regions, but also requires high-level understanding of the image content as well as user intent during photo capture. In this paper, we propose a unified framework to estimate a spatially-varying blur map and understand its desirability in terms of image quality at the same time. In particular, we use a dilated fully convolutional neural network with pyramid pooling and boundary refinement layers to generate high-quality blur response maps. If blur exists, we classify its desirability to three levels ranging from good to bad, by distilling high-level semantics and learning an attention map to adaptively localize the important content in the image. The whole framework is end-to-end jointly trained with both supervisions of pixel-wise blur responses and image-wise blur desirability levels. Considering the limitations of existing image blur datasets, we collected a new large-scale dataset with both annotations to facilitate training. The proposed methods are extensively evaluated on two datasets and demonstrate state-of-the-art performance on both tasks.

Detecting segments of interest from an input sequence is a challenging problem which often requires not only good knowledge of individual target segments, but also contextual understanding of the entire input sequence and the relationships between the target segments. To address this problem, we propose the Sequence-to-Segment Network (S$^2$N), a novel end-to-end sequential encoder-decoder architecture. S$^2$N first encodes the input into a sequence of hidden states that progressively capture both local and holistic information. It then employs a novel decoding architecture, called Segment Detection Unit (SDU), that integrates the decoder state and encoder hidden states to detect segments sequentially. During training, we formulate the assignment of predicted segments to ground truth as bipartite matching and use the Earth Mover's Distance to calculate the localization errors. We experiment with S$^2$N on temporal action proposal generation and video summarization and show that S$^2$N achieves state-of-the-art performance on both tasks.

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.

In this paper, we introduce the problem of simultaneously detecting multiple photographic defects. We aim at detecting the existence, severity, and potential locations of common photographic defects related to color, noise, blur and composition. The automatic detection of such defects could be used to provide users with suggestions for how to improve photos without the need to laboriously try various correction methods. Defect detection could also help users select photos of higher quality while filtering out those with severe defects in photo curation and summarization. To investigate this problem, we collected a large-scale dataset of user annotations on seven common photographic defects, which allows us to evaluate algorithms by measuring their consistency with human judgments. Our new dataset enables us to formulate the problem as a multi-task learning problem and train a multi-column deep convolutional neural network (CNN) to simultaneously predict the severity of all the defects. Unlike some existing single-defect estimation methods that rely on low-level statistics and may fail in many cases on natural photographs, our model is able to understand image contents and quality at a higher level. As a result, in our experiments, we show that our model has predictions with much higher consistency with human judgments than low-level methods as well as several baseline CNN models. Our model also performs better than an average human from our user study.

Computation is a powerful artistic medium. Artists with experience in programming have demonstrated the unique creative opportunities of using code to make art. Currently, manual artists interested in using procedural techniques must undergo the difficult process of learning to program, and must adopt tools and practices far removed from those to which they are accustomed. We hypothesize that, through the right direct manipulation interface, we can enable accessible and expressive procedural art creation. To explore this, we developed Para, a digital illustration tool that supports the creation of declarative constraints in vector artwork. Para's constraints enable procedural relationships while facilitating live manual control and non-linear editing. Constraints can be combined with duplication behaviors and ordered collections of artwork to produce complex, dynamic compositions. We use the results of two open-ended studies with professional artists and designers to provide guidelines for accessible tools that integrate manual and procedural expression.

Automatic image aesthetics rating has received a growing interest with the recent breakthrough in deep learning. Although many studies exist for learning a generic or universal aesthetics model, investigation of aesthetics models incorporating individual user's preference is quite limited. We address this personalized aesthetics problem by showing that individual's aesthetic preferences exhibit strong correlations with content and aesthetic attributes, and hence the deviation of individual's perception from generic image aesthetics is predictable. To accommodate our study, we first collect two distinct datasets, a large image dataset from Flickr and annotated by Amazon Mechanical Turk, and a small dataset of real personal albums rated by owners. We then propose a new approach to personalized aesthetics learning that can be trained even with a small set of annotated images from a user. The approach is based on a residual-based model adaptation scheme which learns an offset to compensate for the generic aesthetics score. Finally, we introduce an active learning algorithm to optimize personalized aesthetics prediction for real-world application scenarios. Experiments demonstrate that our approach can effectively learn personalized aesthetics preferences, and outperforms existing methods on quantitative comparisons.

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.

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.

When creating a photo album of an event, people typically select a few important images to keep or share. There is some consistency in the process of choosing the important images, and discarding the unimportant ones. Modeling this selection process will assist automatic photo selection and album summarization. In this paper, we show that the selection of important images is consistent among different viewers, and that this selection process is related to the event type of the album. We introduce the concept of event-specific image importance. We collected a new event album dataset with human annotation of the relative image importance with each event album. We also propose a Convolutional Neural Network (CNN) based method to predict the image importance score of a given event album, using a novel rank loss function and a progressive training scheme. Results demonstrate that our method significantly outperforms various baseline methods.

We aim at detecting salient objects in unconstrained images. In unconstrained images, the number of salient objects (if any) varies from image to image, and is not given. We present a salient object detection system that directly outputs a compact set of detection windows, if any, for an input image. Our system leverages a Convolutional-Neural-Network model to generate location proposals of salient objects. Location proposals tend to be highly overlapping and noisy. Based on the Maximum a Posteriori principle, we propose a novel subset optimization framework to generate a compact set of detection windows out of noisy proposals. In experiments, we show that our subset optimization formulation greatly enhances the performance of our system, and our system attains 16-34% relative improvement in Average Precision compared with the state-of-the-art on three challenging salient object datasets.

People can immediately and precisely identify that an image contains 1, 2, 3 or 4 items by a simple glance. The phenomenon, known as Subitizing, inspires us to pursue the task of Salient Object Subitizing (SOS), i.e. predicting the existence and the number of salient objects in a scene using holistic cues. To study this problem, we propose a new image dataset annotated using an online crowdsourcing marketplace. We show that a proposed subitizing technique using an end-to-end Convolutional Neural Network (CNN) model achieves significantly better than chance performance in matching human labels on our dataset. It attains 94% accuracy in detecting the existence of salient objects, and 42–82% accuracy (chance is 20%) in predicting the number of salient objects (1, 2, 3, and 4+), without resorting to any object localization process. Finally, we demonstrate the usefulness of the proposed subitizing technique in two computer vision applications: salient object detection and object proposal.

This paper investigates problems of image style, aesthetics, and quality estimation, which require fine-grained details from high-resolution images, utilizing deep neural network training approach. Existing deep convolutional neural networks mostly extracted one patch such as a down-sized crop from each image as a training example. However, one patch may not always well represent the entire image, which may cause ambiguity during training. We propose a deep multi-patch aggregation network training approach, which allows us to train models using multiple patches generated from one image. We achieve this by constructing multiple, shared columns in the neural network and feeding multiple patches to each of the columns. More importantly, we propose two novel network layers (statistics and sorting) to support aggregation of those patches. The proposed deep multi-patch aggregation network integrates shared feature learning and aggregation function learning into a unified framework. We demonstrate the effectiveness of the deep multi-patch aggregation network on the three problems, i.e., image style recognition, aesthetic quality categorization, and image quality estimation. Our models trained using the proposed networks significantly outperformed the state of the art in all three applications.

We propose a highly efficient, yet powerful, salient object detection method based on the Minimum Barrier Distance (MBD) Transform. The MBD transform is robust to pixel-value fluctuation, and thus can be effectively applied on raw pixels without region abstraction. We present an approximate MBD transform algorithm with 100X speedup over the exact algorithm. An error bound analysis is also provided. Powered by this fast MBD transform algorithm, the proposed salient object detection method runs at 80 FPS, and significantly outperforms previous methods with similar speed on four large benchmark datasets, and achieves comparable or better performance than state-of-the-art methods. Furthermore, a technique based on color whitening is proposed to extend our method to leverage the appearance-based backgroundness cue. This extended version further improves the performance, while still being one order of magnitude faster than all the other leading methods.

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.

Cropping is one of the most fundamental and common operations in image processing for improving the aesthetic quality of photographs. Instead of manually designing rules for cropping, in this paper, we propose a generative model that learns an aesthetic photo cropping cascade from a large database of well-composed images and a dataset containing images with crops generated by expert photographers. Specifically, this model includes cropping priori, intuitive likelihood, compositional likelihood and change likelihood. Our learning exploits a spatial pyramid saliency feature and a multi-level foreground segmentation. The inference is done by efficient sub window search (ESS) [10] which is benefited from the bound at conditional distribution in the cascade. Additionally, for extracting attentional subjects and capturing scene composition, we design an iterative saliency method to model the saliency moving paths, which is beyond the typical saliency model predicting a single attentional region. Experiments show that our approach outperforms the state-of-the-art cropping methods by a large margin.

Abstract We present a data‐driven method for automatically cropping photographs to be well‐composed and aesthetically pleasing. Our method matches the composition of an amateur's photograph to an expert's using point correspondences. The correspondences are based on a novel high‐level local descriptor we term the ‘Object Context’. Object Context is an extension of Shape Context: it is a descriptor encoding which objects and scene elements surround a given point. By searching a database of expertly composed images, we can find a crop window which makes an amateur's photograph closely match the composition of a database exemplar. We cull irrelevant matches in the database efficiently using a global descriptor which encodes the objects in the scene. For images with similar content in the database, we efficiently search the space of possible crops using generalized Hough voting. When comparing the result of our algorithm to expert crops, our crop windows overlap the expert crops by 83.6%. We also perform a user study which shows that our crops compare favourably to an expert humans' crops.

Cropping is one of the most common tasks in image editing for improving the aesthetic quality of a photograph. In this paper, we propose a new, aesthetic photo cropping system which combines three models: visual composition, boundary simplicity, and content preservation. The visual composition model measures the quality of composition for a given crop. Instead of manually defining rules or score functions for composition, we learn the model from a large set of well-composed images via discriminative classifier training. The boundary simplicity model measures the clearness of the crop boundary to avoid object cutting-through. The content preservation model computes the amount of salient information kept in the crop to avoid excluding important content. By assigning a hard lower bound constraint on the content preservation and linearly combining the scores from the visual composition and boundary simplicity models, the resulting system achieves significant improvement over recent cropping methods in both quantitative and qualitative evaluation.

Abstract In this work we detail a method that leverages the two color heads of recent low‐end fused deposition modeling (FDM) 3D printers to produce continuous tone imagery. The challenge behind producing such two‐tone imagery is how to finely interleave the two colors while minimizing the switching between print heads, making each color printed span as long and continuous as possible to avoid artifacts associated with printing short segments. The key insight behind our work is that by applying small geometric offsets, tone can be varied without the need to switch color print heads within a single layer. We can now effectively print (two‐tone) texture mapped models capturing both geometric and color information in our output 3D prints.

Abstract Procedural tree models have been popular in computer graphics for their ability to generate a variety of output trees from a set of input parameters and to simulate plant interaction with the environment for a realistic placement of trees in virtual scenes. However, defining such models and their parameters is a difficult task. We propose an inverse modelling approach for stochastic trees that takes polygonal tree models as input and estimates the parameters of a procedural model so that it produces trees similar to the input. Our framework is based on a novel parametric model for tree generation and uses Monte Carlo Markov Chains to find the optimal set of parameters. We demonstrate our approach on a variety of input models obtained from different sources, such as interactive modelling systems, reconstructed scans of real trees and developmental models.

Abstract We propose an optimization framework for 3D printing that seeks to save printing time and the support material required to print 3D shapes. Three‐dimensional printing technology is rapidly maturing and may revolutionize how we manufacture objects. The total cost of printing, however, is governed by numerous factors which include not only the price of the printer but also the amount of material and time to fabricate the shape. Our PackMerger framework converts the input 3D watertight mesh into a shell by hollowing its inner parts. The shell is then divided into segments. The location of splits is controlled based on several parameters, including the size of the connection areas or volume of each segment. The pieces are then tightly packed using optimization. The optimization attempts to minimize the amount of support material and the bounding box volume of the packed segments while keeping the number of segments minimal. The final packed configuration can be printed with substantial time and material savings, while also allowing printing of objects that would not fit into the printer volume. We have tested our system on three different printers and it shows a reduction of 5–30% of the printing time while simultaneously saving 15–65% of the support material. The optimization time was approximately 1 min. Once the segments are printed, they need to be assembled.

Structured decorative patterns are common ornamentations in a variety of media like books, web pages, greeting cards and interior design. Creating such art from scratch using conventional software is time consuming for experts and daunting for novices. We introduce DecoBrush, a data-driven drawing system that generalizes the conventional digital "painting" concept beyond the scope of natural media to allow synthesis of structured decorative patterns following user-sketched paths. The user simply selects an example library and draws the overall shape of a pattern. DecoBrush then synthesizes a shape in the style of the exemplars but roughly matching the overall shape. If the designer wishes to alter the result, DecoBrush also supports user-guided refinement via simple drawing and erasing tools. For a variety of example styles, we demonstrate high-quality user-constrained synthesized patterns that visually resemble the exemplars while exhibiting plausible structural variations.

Existing natural media painting simulations have produced high-quality results, but have required powerful compute hardware and have been limited to screen resolutions. Digital artists would like to be able to use watercolor-like painting tools, but at print resolutions and on lower end hardware such as laptops or even slates. We present a procedural algorithm for generating watercolor-like dynamic paint behaviors in a lightweight manner. Our goal is not to exactly duplicate watercolor painting, but to create a range of dynamic behaviors that allow users to achieve a similar style of process and result, while at the same time having a unique character of its own. Our stroke representation is vector based, allowing for rendering at arbitrary resolutions, and our procedural pigment advection algorithm is fast enough to support painting on slate devices. We demonstrate our technique in a commercially available slate application used by professional artists. Finally, we present a detailed analysis of the different vector-rendering technologies available.

Existing natural media painting simulations have produced high quality results, but have required powerful compute hardware and have been limited to screen resolutions. Digital artists would like to be able to use watercolor-like painting tools, but at print resolutions and on lower end hardware such as laptops or even slates. We present a procedural algorithm for generating watercolor-like dynamic paint behaviors in a lightweight manner. Our goal is not to exactly duplicate watercolor painting, but to create a range of dynamic behaviors that allow users to achieve a similar style of process and result, while at the same time having a unique character of its own. Our stroke representation is vector-based, allowing for rendering at arbitrary resolutions, and our procedural pigment advection algorithm is fast enough to support painting on slate devices. We demonstrate our technique in a commercially available slate application used by professional artists.

Design patterns have proven useful in many creative fields, providing content creators with archetypal, reusable guidelines to leverage in projects. Creating such patterns, however, is a time-consuming, manual process, typically relegated to a few experts in any given domain. In this paper, we describe an algorithmic method for learning design patterns directly from data using techniques from natural language processing and structured concept learning. Given a set of labeled, hierarchical designs as input, we induce a probabilistic formal grammar over these exemplars. Once learned, this grammar encodes a set of generative rules for the class of designs, which can be sampled to synthesize novel artifacts. We demonstrate the method on geometric models and Web pages, and discuss how the learned patterns can drive new interaction mechanisms for content creators.

We present a dynamic tree modeling and representation technique that allows complex tree models to interact with their environment. Our method uses changes in the light distribution and proximity to solid obstacles and other trees as approximations of biologically motivated transformations on a skeletal representation of the tree's main branches and its procedurally generated foliage. Parts of the tree are transformed only when required, thus our approach is much faster than common algorithms such as Open L-Systems or space colonization methods. Input is a skeleton-based tree geometry that can be computed from common tree production systems or from reconstructed laser scanning models. Our approach enables content creators to directly interact with trees and to create visually convincing ecosystems interactively. We present different interaction types and evaluate our method by comparing our transformations to biologically based growth simulation techniques.

The use of 3D printing has rapidly expanded in the past couple of years. It is now possible to produce 3D-printed objects with exceptionally high fidelity and precision. However, although the quality of 3D printing has improved, both the time to print and the material costs have remained high. Moreover, there is no guarantee that a printed model is structurally sound. The printed product often does not survive cleaning, transportation, or handling, or it may even collapse under its own weight. We present a system that addresses this issue by providing automatic detection and correction of the problematic cases. The structural problems are detected by combining a lightweight structural analysis solver with 3D medial axis approximations. After areas with high structural stress are found, the model is corrected by combining three approaches: hollowing, thickening, and strut insertion. Both detection and correction steps are repeated until the problems have been eliminated. Our process is designed to create a model that is visually similar to the original model but possessing greater structural integrity.

Abstract Procedural methods present one of the most powerful techniques for authoring a vast variety of computer graphics models. However, their massive applicability is hindered by the lack of control and a low predictability of the results. In the classical procedural modeling pipeline, the user usually defines a set of rules, executes the procedural system, and by examining the results attempts to infer what should be changed in the system definition in order to achieve the desired output. We present guided procedural modeling, a new approach that allows a high level of top‐down control by breaking the system into smaller building blocks that communicate. In our work we generalize the concept of the environment. The user creates a set of guides. Each guide defines a region in which a specific procedural model operates. These guides are connected by a set of links that serve for message passing between the procedural models attached to each guide. The entire model consists of a set of guides with procedural models, a graph representing their connection, and the method in which the guides interact. The modeling process is performed by modifying each of the described elements. The user can control the high‐level description by editing the guides or manipulate the low‐level description by changing the procedural rules. Changing the connectivity allows the user to create new complex forms in an easy and intuitive way. We show several examples of procedural structures, including an ornamental pattern, a street layout, a bridge, and a model of trees. We also demonstrate interactive examples for quick and intuitive editing using physics‐based mass‐spring system.

Procedural representations provide powerful means for generating complex geometric structures. They are also notoriously difficult to control. In this article, we present an algorithm for controlling grammar-based procedural models. Given a grammar and a high-level specification of the desired production, the algorithm computes a production from the grammar that conforms to the specification. This production is generated by optimizing over the space of possible productions from the grammar. The algorithm supports specifications of many forms, including geometric shapes and analytical objectives. We demonstrate the algorithm on procedural models of trees, cities, buildings, and Mondrian paintings.

Abstract We present an important step towards the solution of the problem of inverse procedural modeling by generating parametric context‐free L‐systems that represent an input 2D model. The L‐system rules efficiently code the regular structures and the parameters represent the properties of the structure transformations. The algorithm takes as input a 2D vector image that is composed of atomic elements, such as curves and poly‐lines. Similar elements are recognized and assigned terminal symbols of an L‐system alphabet. The terminal symbols' position and orientation are pair‐wise compared and the transformations are stored as points in multiple 4D transformation spaces. By careful analysis of the clusters in the transformation spaces, we detect sequences of elements and code them as L‐system rules. The coded elements are then removed from the clusters, the clusters are updated, and then the analysis attempts to code groups of elements in (hierarchies) the same way. The analysis ends with a single group of elements that is coded as an L‐system axiom. We recognize and code branching sequences of linearly translated, scaled, and rotated elements and their hierarchies. The L‐system not only represents the input image, but it can also be used for various editing operations. By changing the L‐system parameters, the image can be randomized, symmetrized, and groups of elements and regular structures can be edited. By changing the terminal and non‐terminal symbols, elements or groups of elements can be replaced.

Abstract Smart deformation and warping tools play an important part in modern day geometric modeling systems. They allow existing content to be stretched or scaled while preserving visually salient information. To date, these techniques have primarily focused on preserving local shape details, not taking into account important global structures such as symmetry and line features. In this work we present a novel framework that can be used to preserve the global structure in images and vector art. Such structures include symmetries and the spatial relations in shapes and line features in an image. Central to our method is a new formulation of preserving structure as an optimization problem. We use novel optimization strategies to achieve the interactive performance required by modern day modeling applications. We demonstrate the effectiveness of our framework by performing structure preservation deformation of images and complex vector art at interactive rates.

Abstract Understanding symmetries and arrangements in existing content is the first step towards providing higher level content aware editing capabilities. Such capabilities may include edits that both preserve existing structure as well as synthesize entirely new structures based on the extracted pattern rules. In this paper we show how to detect regular symmetries and arrangement along curved segments in vector art. We determine individual elements in the art by using the transformation similarity for sequences of sample points on the input curves. Then we detect arrangements of those elements along an arbitrary curved path. We can un‐warp the arrangement path to detect symmetries near the path. We introduce novel applications inform of editing elements that are arranged along a curved path. This includes their sliding along the path, changing of their spacing, or their scale. We also allow the user to brush the elements that the system recognized along new paths.

We present a method for generating realistic models of temperate-climate trees and shrubs. This method is based on the biological hypothesis that the form of a developing tree emerges from a self-organizing process dominated by the competition of buds and branches for light or space, and regulated by internal signaling mechanisms. Simulations of this process robustly generate a wide range of realistic trees and bushes. The generated forms can be controlled with a variety of interactive techniques, including procedural brushes, sketching, and editing operations such as pruning and bending of branches. We illustrate the usefulness and versatility of the proposed method with diverse tree models, forest scenes, animations of tree development, and examples of combined interactive-procedural tree modeling.

Abstract We present a method for synthesizing two dimensional (2D) element arrangements from an example. The main idea is to combine texture synthesis techniques based‐on a local neighborhood comparison and procedural modeling systems based‐on local growth. Given a user‐specified reference pattern, our system analyzes neigh‐borhood information of each element by constructing connectivity. Our synthesis process starts with a single seed and progressively places elements one by one by searching a reference element which has local features that are the most similar to the target place of the synthesized pattern. To support creative design activities, we introduce three types of interaction for controlling global features of the resulting pattern, namely a spray tool, a flow field tool, and a boundary tool. We also introduce a global optimization process that helps to avoid local error concentrations. We illustrate the feasibility of our method by creating several types of 2D patterns.

No abstract available.

Modeling and rendering of natural scenes with thousands of plants poses a number of problems. The terrain must be modeled and plants must be distributed throughout it in a realistic manner, reflecting the interactions of plants with each other and with their environment. Geometric models of individual plants, consistent with their positions within the ecosystem, must be synthesized to populate the scene. The scene, which may consist of billions of primitives, must be rendered efficiently while incorporating the subtleties of lighting in a natural environment. We have developed a system built around a pipeline of tools that address these tasks. The terrain is designed using an interactive graphical editor. Plant distribution is determined by hand (as one would do when designing a garden), by ecosystem simulation, or by a combination of both techniques. Given parametrized procedural models of individual plants, the geometric complexity of the scene is reduced by approximate instancing, in which similar plants, groups of plants, or plant organs are replaced by instances of representative objects before the scene is rendered. The paper includes examples of visually rich scenes synthesized using the system.

Interaction with the environment is a key factor affecting the development of plants and plant ecosystems. In this paper we introduce a modeling framework that makes it possible to simulate and visualize a wide range of interactions at the level of plant architecture. This framework extends the formalism of Lindenmayer systems with constructs needed to model bi-directional information exchange between plants and their environment. We illustrate the proposed framework with models and simulations that capture the development of tree branches limited by collisions, the colonizing growth of clonal plants competing for space in favorable areas, the interaction between roots competing for water in the soil, and the competition within and between trees for access to light. Computer animation and visualization techniques make it possible to better understand the modeled processes and lead to realistic images of plants within their environmental context. CR categories: F.4.2 [Mathematical Logi...

The paper extends Lindenmayer systems in a manner suitable for simulating the interaction between a developing plant and its environment. The formalism is illustrated by modeling the response of trees to pruning, which yields synthetic images of sculptured plants found in topiary gardens.