Search

Sketch-based modeling strives to bring the ease and immediacy of drawing to the 3D world. However, while drawings are easy for humans to create, they are very challenging for computers to interpret due to their sparsity and ambiguity. We propose a data-driven approach that tackles this challenge by learning to reconstruct 3D shapes from one or more drawings. At the core of our approach is a deep convolutional neural network (CNN) that predicts occupancy of a voxel grid from a line drawing. This CNN provides us with an initial 3D reconstruction as soon as the user completes a single drawing of the desired shape. We complement this single-view network with an updater CNN that refines an existing prediction given a new drawing of the shape created from a novel viewpoint. A key advantage of our approach is that we can apply the updater iteratively to fuse information from an arbitrary number of viewpoints, without requiring explicit stroke correspondences between the drawings. We train both CNNs by rendering synthetic contour drawings from hand-modeled shape collections as well as from procedurally-generated abstract shapes. Finally, we integrate our CNNs in a minimal modeling interface that allows users to seamlessly draw an object, rotate it to see its 3D reconstruction, and refine it by re-drawing from another vantage point using the 3D reconstruction as guidance. The main strengths of our approach are its robustness to freehand bitmap drawings, its ability to adapt to different object categories, and the continuum it offers between single-view and multi-view sketch-based modeling., Comment: See our accompanying video on https://youtu.be/DGIYzmlm2pQ, networks and databases available at https://ns.inria.fr/d3/3DSketching/. To appear in PACMCGIT

We introduce a neural relighting algorithm for captured indoors scenes, that allows interactive free-viewpoint navigation. Our method allows illumination to be changed synthetically, while coherently rendering cast shadows and complex glossy materials. We start with multiple images of the scene and a three-dimensional mesh obtained by multi-view stereo (MVS) reconstruction. We assume that lighting is well explained as the sum of a view-independent diffuse component and a view-dependent glossy term concentrated around the mirror reflection direction. We design a convolutional network around input feature maps that facilitate learning of an implicit representation of scene materials and illumination, enabling both relighting and free-viewpoint navigation. We generate these input maps by exploiting the best elements of both image-based and physically based rendering. We sample the input views to estimate diffuse scene irradiance, and compute the new illumination caused by user-specified light sources using path tracing. To facilitate the network's understanding of materials and synthesize plausible glossy reflections, we reproject the views and compute mirror images . We train the network on a synthetic dataset where each scene is also reconstructed with MVS. We show results of our algorithm relighting real indoor scenes and performing free-viewpoint navigation with complex and realistic glossy reflections, which so far remained out of reach for view-synthesis techniques.

International audience; We present a novel method for authoring landscapes with flora and fauna while considering their mutual interactions. Our algorithm outputs a steadystate ecosystem in the form of density maps for each species, their daily circuits, and a modified terrain with eroded trails from a terrain, climatic conditions, and species with related biological information. We introduce the Resource Access Graph, a new data structure that encodes both interactions between food chain levels and animals traveling between resources over the terrain. A novel competition algorithm operating on this data progressively computes a steady-state solution up the food chain, from plants to carnivores. The user can explore the resulting landscape, where plants and animals are instantiated on the fly, and interactively edit it by over-painting the maps. Our results show that our system enables the authoring of consistent landscapes where the impact of wildlife is visible through animated animals, clearings in the vegetation, and eroded trails. We provide quantitative validation with existing ecosystems and a user-study with expert paleontologist end-users, showing that our system enables them to author and compare different ecosystems illustrating climate changes over the same terrain while enabling relevant visual immersion into consistent landscapes.

There has recently been great interest in neural rendering methods. Some approaches use 3D geometry reconstructed with Multi-View Stereo (MVS) but cannot recover from the errors of this process, while others directly learn a volumetric neural representation, but suffer from expensive training and inference. We introduce a general approach that is initialized with MVS, but allows further optimization of scene properties in the space of input views, including depth and reprojected features, resulting in improved novel-view synthesis. A key element of our approach is our new differentiable point-based pipeline, based on bi-directional Elliptical Weighted Average splatting, a probabilistic depth test and effective camera selection. We use these elements together in our neural renderer, that outperforms all previous methods both in quality and speed in almost all scenes we tested. Our pipeline can be applied to multi-view harmonization and stylization in addition to novel-view synthesis., https://repo-sam.inria.fr/fungraph/differentiable-multi-view/

International audience; Image-based rendering (IBR) provides a rich toolset for free-viewpoint navigation in captured scenes. Many methods exist, usually with an emphasis either on image quality or rendering speed. In this paper we identify common IBR artifacts and combine the strengths of different algorithms to strike a good balance in the speed/quality tradeoff. First, we address the problem of visible color seams that arise from blending casually-captured input images by explicitly treating view-dependent effects. Second, we compensate for geometric reconstruction errors by refining per-view information using a novel clustering and filtering approach. Finally, we devise a practical hybrid IBR algorithm, which locally identifies and utilizes the rendering method best suited for an image region while retaining interactive rates. We compare our method against classical and modern (neural) approaches in indoor and outdoor scenes and demonstrate superiority in quality and/or speed.

International audience; We present CASSIE, a conceptual modeling system in VR that leverages freehand mid-air sketching, and a novel 3D optimization framework to create connected curve network armatures, predictively surfaced using patches with C0 continuity. Our system provides a judicious balance of interactivity and automation, providing a homogeneous 3D drawing interface for a mix of freehand curves, curve networks, and surface patches. Our system encourages and aids users in drawing consistent networks of curves, easing the transition from freehand ideation to concept modeling. A comprehensive user study with professional designers as well as amateurs (N=12), and a diverse gallery of 3D models, show our armature and patch functionality to offer a user experience and expressivity on par with freehand ideation, while creating sophisticated concept models for downstream applications.

International audience; Fashion design often starts with hand-drawn, expressive sketches that communicate the essence of a garment over idealized human bodies. We propose an approach to automatically dress virtual characters from such input, previously complemented with user-annotations. In contrast to prior work requiring users to draw garments with accurate proportions over each virtual character to be dressed, our method follows a style transfer strategy : the information extracted from a single, annotated fashion sketch can be used to inform the synthesis of one to many new garment(s) with similar style, yet different proportions. In particular, we define the style of a loose garment from its silhouette and folds, which we extract from the drawing. Key to our method is our strategy to extract both shape and repetitive patterns of folds from the 2D input. As our results show, each input sketch can be used to dress a variety of characters of different morphologies, from virtual humans to cartoon-style characters.

International audience; Image-Based Rendering allows users to easily capture a scene using a single camera and then navigate freely with realistic results. However, the resulting renderings are completely static, and dynamic effects – such as fire, waterfalls or small waves – cannot be reproduced. We tackle the challenging problem of enabling free-viewpoint navigation including such stationary dynamic effects, but still maintaining the simplicity of casual capture. Using a single camera – instead of previous complex synchronized multi-camera setups – means that we have unsynchronized videos of the dynamic effect from multiple views, making it hard to blend them when synthesizing novel views. We present a solution that allows smooth free-viewpoint video-based rendering (VBR) of such scenes using temporal Laplacian pyramid decomposition video, enabling spatio-temporal blending. For effects such as fire and waterfalls, that are semi-transparent and occupy 3D space, we first estimate their spatial volume. This allows us to create per-video geometries and alpha-matte videos that we can blend using our frequency-dependent method. We also extend Laplacian blending to the temporal dimension to remove additional temporal seams. We show results on scenes containing fire, waterfalls or rippling waves at the seaside, bringing these scenes to life.

International audience; Recent rendering advances dramatically reduce the cost of global illumination. But even with hardware acceleration, complex light paths with multiple glossy interactions are still expensive; our new algorithm stores these paths in precomputed light probes and reprojects them at runtime to provide interactivity. Combined with traditional light maps for diffuse lighting our approach interactively renders all light paths in static scenes with opaque objects. Naively reprojecting probes with glossy lighting is memory-intensive, requires efficient access to the correctly reflected radiance, and exhibits problems at occlusion boundaries in glossy reflections. Our solution addresses all these issues. To minimize memory, we introduce an adaptive light probe parameterization that allocates increased resolution for shinier surfaces and regions of higher geometric complexity. To efficiently sample glossy paths, our novel gathering algorithm reprojects probe texels in a view-dependent manner using efficient reflection estimation and a fast rasterization-based search. Naive probe reprojection often sharpens glossy reflections at occlusion boundaries, due to changes in parallax. To avoid this, we split the convolution induced by the BRDF into two steps: we precompute probes using a lower material roughness and apply an adaptive bilateral filter at runtime to reproduce the original surface roughness. Combining these elements, our algorithm interactively renders complex scenes while fitting in the memory, bandwidth, and computation constraints of current hardware.

We present a sketch-based CAD modeling system, where users create objects incrementally by sketching the desired shape edits, which our system automatically translates to CAD operations. Our approach is motivated by the close similarities between the steps industrial designers follow to draw 3D shapes, and the operations CAD modeling systems offer to create similar shapes. To overcome the strong ambiguity with parsing 2D sketches, we observe that in a sketching sequence, each step makes sense and can be interpreted in the context of what has been drawn before. In our system, this context corresponds to a partial CAD model, inferred in the previous steps, which we feed along with the input sketch to a deep neural network in charge of interpreting how the model should be modified by that sketch. Our deep network architecture then recognizes the intended CAD operation and segments the sketch accordingly, such that a subsequent optimization estimates the parameters of the operation that best fit the segmented sketch strokes. Since there exists no datasets of paired sketching and CAD modeling sequences, we train our system by generating synthetic sequences of CAD operations that we render as line drawings. We present a proof of concept realization of our algorithm supporting four frequently used CAD operations. Using our system, participants are able to quickly model a large and diverse set of objects, demonstrating Sketch2CAD to be an alternate way of interacting with current CAD modeling systems.

International audience; We present the first algorithm capable of automatically lifting real-world, vector-format, industrial design sketches into 3D. Targeting real-world sketches raises numerous challenges due to inaccuracies, use of overdrawn strokes, and construction lines. In particular, while construction lines convey important 3D information, they add significant clutter and introduce multiple accidental 2D intersections. Our algorithm exploits the geometric cues provided by the construction lines and lifts them to 3D by computing their intended 3D intersections and depths. Once lifted to 3D, these lines provide valuable geometric constraints that we leverage to infer the 3D shape of other artist drawn strokes. The core challenge we address is inferring the 3D connectivity of construction and other lines from their 2D projections by separating 2D intersections into 3D intersections and accidental occlusions. We efficiently address this complex combinatorial problem using a dedicated search algorithm that leverages observations about designer drawing preferences , and uses those to explore only the most likely solutions of the 3D intersection detection problem. We demonstrate that our separator outputs are of comparable quality to human annotations, and that the 3D structures we recover enable a range of design editing and visualization applications, including novel view synthesis and 3D-aware scaling of the depicted shape.

International audience; Multi-view stereo can be used to rapidly create realistic virtual content, such as textured meshes or a geometric proxy for free-viewpoint Image-Based Rendering (IBR). These solutions greatly simplify the content creation process compared to traditional methods, but it is difficult to modify the content of the scene. We propose a novel approach to create scenes by composing (parts of) multiple captured scenes. The main difficulty of such compositions is that lighting conditions in each captured scene are different; to obtain a realistic composition we need to make lighting coherent. We propose a two-pass solution, by adapting a multi-view relighting network. We first match the lighting conditions of each scene separately and then synthesize shadows between scenes in a subsequent pass. We also improve the realism of the composition by estimating the change in ambient occlusion in contact areas between parts and compensate for the color balance of the different cameras used for capture. We illustrate our method with results on multiple compositions of outdoor scenes and show its application to multi-view image composition, IBR and textured mesh creation.

Performance is a critical challenge in mobile image processing. Given a reference imaging pipeline, or even human-adjusted pairs of images, we seek to reproduce the enhancements and enable real-time evaluation. For this, we introduce a new neural network architecture inspired by bilateral grid processing and local affine color transforms. Using pairs of input/output images, we train a convolutional neural network to predict the coefficients of a locally-affine model in bilateral space. Our architecture learns to make local, global, and content-dependent decisions to approximate the desired image transformation. At runtime, the neural network consumes a low-resolution version of the input image, produces a set of affine transformations in bilateral space, upsamples those transformations in an edge-preserving fashion using a new slicing node, and then applies those upsampled transformations to the full-resolution image. Our algorithm processes high-resolution images on a smartphone in milliseconds, provides a real-time viewfinder at 1080p resolution, and matches the quality of state-of-the-art approximation techniques on a large class of image operators. Unlike previous work, our model is trained off-line from data and therefore does not require access to the original operator at runtime. This allows our model to learn complex, scene-dependent transformations for which no reference implementation is available, such as the photographic edits of a human retoucher., 12 pages, 14 figures, Siggraph 2017

International audience; Texture, highlights, and shading are some of many visual cues that allow humans to perceive material appearance in pictures. Designing algorithms able to leverage these cues to recover spatially-varying bi-directional reflectance distribution functions (SVBRDFs) from a few images has challenged computer graphics researchers for decades. I explore the use of deep learning to tackle lightweight appearance capture and make sense of these visual cues. Our networks are capable of recovering per-pixel normals, diffuse albedo, specular albedo and specular roughness from as little as one picture of a flat surface lit by a hand-held flash. We propose a method which improves its prediction with the number of input pictures, and reaches high quality reconstructions with up to 10 images -- a sweet spot between existing single-image and complex multi-image approaches. We introduce several innovations on training data acquisition and network design, bringing clear improvement over the state of the art for lightweight material capture.

In stereoscopic displays, such as those used in VR/AR headsets, our eyes are presented with two different views. The disparity between the views is typically used to convey depth cues, but it could be also used to enhance image appearance. We devise a novel technique that takes advantage of binocular fusion to boost perceived local contrast and visual quality of images. Since the technique is based on fixed tone curves, it has negligible computational cost and it is well suited for real-time applications, such as VR rendering. To control the trade-off between contrast gain and binocular rivalry, we conduct a series of experiments to explain the factors that dominate rivalry perception in a dichoptic presentation where two images of different contrasts are displayed. With this new finding, we can effectively enhance contrast and control rivalry in mono- and stereoscopic images, and in VR rendering, as confirmed in validation experiments.

Empowered by deep learning, recent methods for material capture can estimate a spatially-varying reflectance from a single photograph. Such lightweight capture is in stark contrast with the tens or hundreds of pictures required by traditional optimization-based approaches. However, a single image is often simply not enough to observe the rich appearance of real-world materials. We present a deep-learning method capable of estimating material appearance from a variable number of uncalibrated and unordered pictures captured with a handheld camera and flash. Thanks to an order-independent fusing layer, this architecture extracts the most useful information from each picture, while benefiting from strong priors learned from data. The method can handle both view and light direction variation without calibration. We show how our method improves its prediction with the number of input pictures, and reaches high quality reconstructions with as little as 1 to 10 images -- a sweet spot between existing single-image and complex multi-image approaches., Comment: Accepted to EGSR 2019 in the CGF track

International audience; Recent works on data-driven sketch-based modeling use either voxel grids or normal/depth maps as geometric representations compatible with convolutional neural networks. While voxel grids can represent complete objects-including parts not visible in the sketches-their memory consumption restricts them to low-resolution predictions. In contrast, a single normal or depth map can capture fine details, but multiple maps from different viewpoints need to be predicted and fused to produce a closed surface. We propose to combine these two representations to address their respective shortcomings in the context of a multi-view sketch-based modeling system. Our method predicts a voxel grid common to all the input sketches, along with one normal map per sketch. We then use the voxel grid as a support for normal map fusion by optimizing its extracted surface such that it is consistent with the re-projected normals, while being as piecewise-smooth as possible overall. We compare our method with a recent voxel prediction system, demonstrating improved recovery of sharp features over a variety of man-made objects.

International audience; We introduce Delaunay Point Processes, a framework for the extraction of geometric structures from images. Our approach simultaneously locates and groups geometric primitives (line segments, triangles) to form extended structures (line networks, polygons) for a variety of image analysis tasks. Similarly to traditional point processes, our approach uses Markov Chain Monte Carlo to minimize an energy that balances fidelity to the input image data with geometric priors on the output structures. However, while existing point processes struggle to model structures composed of interconnected components, we propose to embed the point process into a Delaunay triangulation, which provides high-quality connectivity by construction. We further leverage key properties of the Delaunay triangulation to devise a fast Markov Chain Monte Carlo sampler. We demonstrate the flexibility of our approach on a variety of applications, including line network extraction, object contouring, and mesh-based image compression.

We propose the first learning-based algorithm that can relight images in a plausible and controllable manner given multiple views of an outdoor scene. In particular, we introduce a geometry-aware neural network that utilizes multiple geometry cues (normal maps, specular direction, etc.) and source and target shadow masks computed from a noisy proxy geometry obtained by multi-view stereo. Our model is a three-stage pipeline: two subnetworks refine the source and target shadow masks, and a third performs the final relighting. Furthermore, we introduce a novel representation for the shadow masks, which we call RGB shadow images. They reproject the colors from all views into the shadowed pixels and enable our network to cope with inacuraccies in the proxy and the non-locality of the shadow casting interactions. Acquiring large-scale multi-view relighting datasets for real scenes is challenging, so we train our network on photorealistic synthetic data. At train time, we also compute a noisy stereo-based geometric proxy, this time from the synthetic renderings. This allows us to bridge the gap between the real and synthetic domains. Our model generalizes well to real scenes. It can alter the illumination of drone footage, image-based renderings, textured mesh reconstructions, and even internet photo collections.

International audience; 3D modeling remains a notoriously difficult task for novices despite significant research effort to provide intuitive and automated systems. We tackle this problem by combining the strengths of two popular domains: sketch-based modeling and procedural modeling. On the one hand, sketch-based modeling exploits our ability to draw but requires detailed, unambiguous drawings to achieve complex models. On the other hand, procedural modeling automates the creation of precise and detailed geometry but requires the tedious definition and parameterization of procedural models. Our system uses a collection of simple procedural grammars, called snippets, as building blocks to turn sketches into realistic 3D models. We use a machine learning approach to solve the inverse problem of finding the procedural model that best explains a user sketch. We use non-photorealistic rendering to generate artificial data for training con-volutional neural networks capable of quickly recognizing the procedural rule intended by a sketch and estimating its parameters. We integrate our algorithm in a coarse-to-fine urban modeling system that allows users to create rich buildings by successively sketching the building mass, roof, facades, windows, and ornaments. A user study shows that by using our approach non-expert users can generate complex buildings in just a few minutes.

International audience; Image-Based Rendering (IBR) allows high-fidelity free-viewpoint navigation using only a set of photographs and 3D reconstruction as input. It is often necessary or convenient to remove objects from the captured scenes, allowing a form of scene editing for IBR. This requires multi-view inpainting of the input images. Previous methods suffer from several major limitations: they lack true multi-view coherence, resulting in artifacts such as blur, they do not preserve perspective during inpainting, provide inaccurate depth completion and can only handle scenes with a few tens of images. Our approach addresses these limitations by introducing a new multi-view method that performs inpainting in intermediate, locally common planes. Use of these planes results in correct perspective and multi-view coherence of inpainting results. For efficient treatment of large scenes, we present a fast planar region extraction method operating on small image clusters. We adapt the resolution of inpainting to that required in each input image of the multi-view dataset, and carefully handle image resampling between the input images and rectified planes. We show results on large indoors and outdoors environments.

Texture, highlights, and shading are some of many visual cues that allow humans to perceive material appearance in single pictures. Yet, recovering spatially-varying bi-directional reflectance distribution functions (SVBRDFs) from a single image based on such cues has challenged researchers in computer graphics for decades. We tackle lightweight appearance capture by training a deep neural network to automatically extract and make sense of these visual cues. Once trained, our network is capable of recovering per-pixel normal, diffuse albedo, specular albedo and specular roughness from a single picture of a flat surface lit by a hand-held flash. We achieve this goal by introducing several innovations on training data acquisition and network design. For training, we leverage a large dataset of artist-created, procedural SVBRDFs which we sample and render under multiple lighting directions. We further amplify the data by material mixing to cover a wide diversity of shading effects, which allows our network to work across many material classes. Motivated by the observation that distant regions of a material sample often offer complementary visual cues, we design a network that combines an encoder-decoder convolutional track for local feature extraction with a fully-connected track for global feature extraction and propagation. Many important material effects are view-dependent, and as such ambiguous when observed in a single image. We tackle this challenge by defining the loss as a differentiable SVBRDF similarity metric that compares the renderings of the predicted maps against renderings of the ground truth from several lighting and viewing directions. Combined together, these novel ingredients bring clear improvement over state of the art methods for single-shot capture of spatially varying BRDFs., Comment: 15 pages, presented at Siggraph 2018

International audience; Recent color transfer methods use local information to learn the transformation from a source to an exemplar image, and then transfer this appearance change to a target image. These solutions achieve very successful results for general mood changes, e.g., changing the appearance of an image from ``sunny'' to ``overcast''. However, such methods have a hard time creating new image content, such as leaves on a bare tree. Texture transfer, on the other hand, can synthesize such content but tends to destroy image structure. We propose the first algorithm that unifies color and texture transfer, outperforming both by leveraging their respective strengths. A key novelty in our approach resides in teasing apart appearance changes that can be modeled simply as changes in color versus those that require new image content to be generated. Our method starts with an analysis phase which evaluates the success of color transfer by comparing the exemplar with the source. This analysis then drives a selective, iterative texture transfer algorithm that simultaneously predicts the success of color transfer on the target and synthesizes new content where needed. We demonstrate our unified algorithm by transferring large temporal changes between photographs, such as change of season -- e.g., leaves on bare trees or piles of snow on a street -- and flooding.

International audience; Developable materials are ubiquitous in design and manufacturing. Unfortunately, general-purpose modeling tools are not suited to modeling 3D objects composed of developable parts. We propose an interactive tool to model such objects from a photograph. Users of our system load a single picture of the object they wish to model, which they annotate to indicate silhouettes and part boundaries. Assuming that the object is symmetric, we also ask users to provide a few annotations of symmetric correspondences. The object is then automatically reconstructed in 3D. At the core of our method is an algorithm to infer the 2D projection of rulings of a developable surface from the traced silhouettes and boundaries. We impose that the surface normal is constant along each ruling, which is a necessary property for the surface to be developable. We complement these developability constraints with symmetry constraints to lift the curve network in 3D. In addition to a 3D model, we output 2D patterns enabling to fabricate real prototypes of the object on the photo. This makes our method well suited for reverse engineering products made of leather, bent cardboard or metal sheets.

International audience; Head-mounted displays (HMDs) often cause discomfort and even nausea. Improving comfort is therefore one of the most significant challenges for the design of such systems. In this paper, we evaluate the effect of different HMD display configurations on discomfort. We do this by designing a device to measure human visual behavior and evaluate viewer comfort. In particular, we focus on one known source of discomfort: the vergence-accommodation (VA) conflict. The VA conflict is the difference between accommodative and vergence response. In HMDs the eyes accommodate to a fixed screen distance while they converge to the simulated distance of the object of interest, requiring the viewer to undo the neural coupling between the two responses. Several methods have been proposed to alleviate the VA conflict, including Depth-of-Field (DoF) rendering, focus-adjustable lenses, and monovision. However, no previous work has investigated whether these solutions actually drive accommodation to the distance of the simulated object. If they did, the VA conflict would disappear, and we expect comfort to improve. We design the first device that allows us to measure accommodation in HMDs, and we use it to obtain accommodation measurements and to conduct a discomfort study. The results of the first experiment demonstrate that only the focus-adjustable-lens design drives accommodation effectively, while other solutions do not drive accommodation to the simulated distance and thus do not resolve the VA conflict. The second experiment measures discomfort. The results validate that the focus-adjustable-lens design improves comfort significantly more than the other solutions.

International audience; Our aim is to give users real-time free-viewpoint rendering of real indoor scenes, captured with off-the-shelf equipment such as a high-quality color camera and a commodity depth sensor. Image-based Rendering (IBR) can provide the realistic imagery required at real-time speed. For indoor scenes however, two challenges are especially prominent. First, the reconstructed 3D geometry must be compact, but faithful enough to respect occlusion relationships when viewed up close. Second, man-made materials call for view-dependent tex-turing, but using too many input photographs reduces performance. We customize a typical RGB-D 3D surface reconstruction pipeline to produce a coarse global 3D surface, and local, per-view geometry for each input image. Our tiled IBR preserves quality by economizing on the expected contributions that entire groups of input pixels make to a final image. The two components are designed to work together, giving real-time performance, while hardly sacrificing quality. Testing on a variety of challenging scenes shows that our inside-out IBR scales favorably with the number of input images.

International audience; We address the problem of multi-view video segmentation of dynamic scenes in general and outdoor environments with possibly moving cameras. Multi-view methods for dynamic scenes usually rely on geometric calibration to impose spatial shape constraints between viewpoints. In this paper, we show that the calibration constraint can be relaxed while still getting competitive segmentation results using multi-view constraints. We introduce new multi-view cotemporality constraints through motion correlation cues, in addition to common appearance features used by co-segmentation methods to identify co-instances of objects. We also take advantage of learning based segmentation strategies by casting the problem as the selection of monocular proposals that satisfy multi-view constraints. This yields a fully automated method that can segment subjects of interest without any particular pre-processing stage. Results on several challenging outdoor datasets demonstrate the feasibility and robustness of our approach.

International audience; We propose a method to remove objects such as people and cars from multi-view urban image datasets, enabling free-viewpoint Image-Based Rendering (IBR) in the edited scenes. Our method combines information from multi-view 3D reconstruction with image inpainting techniques, by formulating the problem as an optimization of a global patch-based objective function. We use IBR techniques to reproject information from neighboring views, and 3D multi-view stereo reconstruction to perform multi-view coherent initialization for inpainting of pixels not filled by reprojection. Our algorithm performs multi-view consistent inpainting for color and 3D by blending reprojections with patch-based image inpaint-ing. We run our algorithm on casually captured datasets, and Google Street View data, removing objects such as cars, people and pillars, showing that our approach produces results of sufficient quality for free-viewpoint IBR on " cleaned up " scenes, as well as IBR scene editing, such as limited displacement of real objects.

Summarization: Comfortable, high-quality 3D stereo viewing is becoming a requirement for interactive applications today. Previous research shows that manipulating disparity can alleviate some of the discomfort caused by 3D stereo, but it is best to do this locally, around the object the user is gazing at. The main challenge is thus to develop a gaze predictor in the demanding context of real-time, heavily task-oriented applications such as games. Our key observation is that player actions are highly correlated with the present state of a game, encoded by game variables. Based on this, we train a classifier to learn these correlations using an eye-tracker which provides the ground-truth object being looked at. The classifier is used at runtime to predict object category - and thus gaze - during game play, based on the current state of game variables. We use this prediction to propose a dynamic disparity manipulation method, which provides rich and comfortable depth. We evaluate the quality of our gaze predictor numerically and experimentally, showing that it predicts gaze more accurately than previous approaches. A subjective rating study demonstrates that our localized disparity manipulation is preferred over previous methods. Παρουσιάστηκε στο: 18th IEEE Virtual Reality Conference

International audience; Image-Based Rendering (IBR) algorithms generate highquality photo-realistic imagery without the burden of detailedmodeling and expensive realistic rendering. Recentmethods have different strengths and weaknesses, dependingon 3D reconstruction quality and scene content. Eachalgorithm operates with a set of hypotheses about the sceneand the novel views, resulting in different quality/speedtrade-offs in different image regions. We present a principledapproach to select the algorithm with the best quality/speedtrade-off in each region. To do this, we proposea Bayesian approach, modeling the rendering quality, therendering process and the validity of the assumptions ofeach algorithm. We then choose the algorithm to use withMaximum a Posteriori estimation. We demonstrate the utilityof our approach on recent IBR algorithms which useoversegmentation and are based on planar reprojection andshape-preserving warps respectively. Our algorithm selectsthe best rendering algorithm for each superpixel in apreprocessing step; at runtime our selective IBR uses thischoice to achieve significant speedup at equivalent or betterquality compared to previous algorithms.

International audience; Many design tasks involve the creation of new objects in the context of an existing scene. Existing work in computer vision only provides partial support for such tasks. On the one hand, multi-view stereo algorithms allow the reconstruction of real-world scenes, while on the other hand algorithms for line-drawing interpretation do not take context into account. Our work combines the strength of these two domains to interpret line drawings of imaginary objects drawn over photographs of an existing scene. The main challenge we face is to identify the existing 3D structure that correlates with the line drawing while also allowing the creation of new structure that is not present in the real world. We propose a labeling algorithm to tackle this problem , where some of the labels capture dominant orientations of the real scene while a free label allows the discovery of new orientations in the imaginary scene. We illustrate our algorithm by interpreting line drawings for urban planing, home remodeling, furniture design and cultural heritage.

International audience; View-dependent surface color of virtual objects can be represented by outgoing radiance of the surface. In this paper we tackle the processing of outgoing radiance stored as a vertex attribute of triangle meshes. Data resulting from an acquisition process can be very large and computationally intensive to render. We show that when reducing the global memory footprint of such acquired objects, smartly reducing the spatial resolution is an effective strategy for overall appearance preservation. Whereas state-of-the-art simplification processes only consider scalar or vectorial attributes , we conversely consider radiance functions defined on the surface for which we derive a metric. For this purpose, several tools are introduced like coherent radiance function interpolation, gradient computation , and distance measurements. Both synthetic and acquired examples illustrate the benefit and the relevance of this radiance-aware simplification process.

International audience; IBR Algorithm Real surface Reconstruction Fig. 1. Image-based rendering blends contributions from different input images to synthesize a novel view. (a) This blending operation is complex for a variety of reasons. For example, incorrect visibility (e.g., for the non-reconstructed green surface in input view (1)) may result in wrong projections of an image into the novel view. Blending must also account for, e.g., differences in projected resolution ((2) and (3)). Previous methods have used hand-crafted heuristics to overcome these issues. (b) Our method generates a set of ranked contributions (mosaics) from the input images and uses predicted blending weights from a CNN to perform deep blending and synthesize (c) novel views. Our solution significantly reduces visible artifacts compared to (d) previous methods. Free-viewpoint image-based rendering (IBR) is a standing challenge. IBR methods combine warped versions of input photos to synthesize a novel view. The image quality of this combination is directly affected by geometric inaccuracies of multi-view stereo (MVS) reconstruction and by view-and image-dependent effects that produce artifacts when contributions from different input views are blended. We present a new deep learning approach to blending for IBR, in which we use held-out real image data to learn blending weights to combine input photo contributions. Our Deep Blending method requires us to address several challenges to achieve our goal of interactive free-viewpoint IBR navigation. We first need to provide sufficiently accurate geometry so the Convolutional Neural Network (CNN) can succeed in finding correct blending weights. We do this by combining two different MVS reconstructions with complementary accuracy vs. completeness tradeoffs. To tightly integrate learning in an interactive IBR system, we need to adapt our rendering algorithm to produce a fixed number of input layers that can then be blended by the CNN. We generate training data with a variety of captured scenes, using each input photo as ground truth in a held-out approach. We also design the network architecture and the training loss to provide high quality novel view synthesis, while reducing temporal flickering artifacts. Our results demonstrate free-viewpoint IBR in a wide variety of scenes, clearly surpassing previous methods in visual quality, especially when moving far from the input cameras.

International audience; Path tracing is now the standard method used to generate realistic imagery in many domains, e.g., film, special effects, architecture etc. Path guiding has recently emerged as a powerful strategy to counter the notoriously long computation times required torender such images. We present a practical path guiding algorithm that performs product sampling, i.e., samples proportionalto the product of the bidirectional scattering distribution function (BSDF) and incoming radiance. We use a spatial-directionalsubdivision to represent incoming radiance, and introduce the use of Linearly Transformed Cosines (LTCs) to represent theBSDF during path guiding, thus enabling efficient product sampling. Despite the computational efficiency of LTCs, several optimizations are needed to make our method cost effective. In particular, we show how we can use vectorization, precomputation,as well as strategies to optimize multiple importance sampling and Russian roulette to improve performance. We evaluate ourmethod on several scenes, demonstrating consistent improvement in efficiency compared to previous work, especially in sceneswith significant glossy inter-reflection.