Yasutaka Furukawa

Building Rome in a day

We present a system that can match and reconstruct 3D scenes from extremely large collections of photographs such as those found by searching for a given city (e.g., Rome) on Internet photo sharing sites. Our system uses a collection of novel parallel distributed matching and reconstruction algorithms, designed to maximize parallelism at each stage in the pipeline and minimize serialization bottlenecks. It is designed to scale gracefully with both the size of the problem and the amount of available computation. We have experimented with a variety of alternative algorithms at each stage of the pipeline and report on which ones work best in a parallel computing environment. Our experimental results demonstrate that it is now possible to reconstruct cities consisting of 150K images in less than a day on a cluster with 500 compute cores.

Towards Internet-scale multi-view stereo

This paper introduces an approach for enabling existing multi-view stereo methods to operate on extremely large unstructured photo collections. The main idea is to decompose the collection into a set of overlapping sets of photos that can be processed in parallel, and to merge the resulting reconstructions. This overlapping clustering problem is formulated as a constrained optimization and solved iteratively. The merging algorithm, designed to be parallel and out-of-core, incorporates robust filtering steps to eliminate low-quality reconstructions and enforce global visibility constraints. The approach has been tested on several large datasets downloaded from Flickr.com, including one with over ten thousand images, yielding a 3D reconstruction with nearly thirty million points.

Reconstructing building interiors from images

This paper proposes a fully automated 3D reconstruction and visualization system for architectural scenes (interiors and exteriors). The reconstruction of indoor environments from photographs is particularly challenging due to texture-poor planar surfaces such as uniformly-painted walls. Our system first uses structure-from-motion, multi-view stereo, and a stereo algorithm specifically designed for Manhattan-world scenes (scenes consisting predominantly of piece-wise planar surfaces with dominant directions) to calibrate the cameras and to recover initial 3D geometry in the form of oriented points and depth maps. Next, the initial geometry is fused into a 3D model with a novel depth-map integration algorithm that, again, makes use of Manhattan-world assumptions and produces simplified 3D models. Finally, the system enables the exploration of reconstructed environments with an interactive, image-based 3D viewer. We demonstrate results on several challenging datasets, including a 3D reconstruction and image-based walk-through of an entire floor of a house, the first result of this kind from an automated computer vision system.

Manhattan-world stereo

Multi-view stereo (MVS) algorithms now produce reconstructions that rival laser range scanner accuracy. However, stereo algorithms require textured surfaces, and therefore work poorly for many architectural scenes (e.g., building interiors with textureless, painted walls). This paper presents a novel MVS approach to overcome these limitations for Manhattan World scenes, i.e., scenes that consists of piece-wise planar surfaces with dominant directions. Given a set of calibrated photographs, we first reconstruct textured regions using an existing MVS algorithm, then extract dominant plane directions, generate plane hypotheses, and recover per-view depth maps using Markov random fields. We have tested our algorithm on several datasets ranging from office interiors to outdoor buildings, and demonstrate results that outperform the current state of the art for such texture-poor scenes.

Accurate Camera Calibration from Multi-View Stereo and Bundle Adjustment

The advent of high-resolution digital cameras and sophisticated multi-view stereo algorithms offers the promise of unprecedented geometric fidelity in image-based modeling tasks, but it also puts unprecedented demands on camera calibration to fulfill these promises. This paper presents a novel approach to camera calibration where top-down information from rough camera parameter estimates and the output of a multi-view-stereo system on scaled-down input images is used to effectively guide the search for additional image correspondences and significantly improve camera calibration parameters using a standard bundle adjustment algorithm (Lourakis and Argyros 2008). The proposed method has been tested on six real datasets including objects without salient features for which image correspondences cannot be found in a purely bottom-up fashion, and objects with high curvature and thin structures that are lost in visual hull construction even with small errors in camera parameters. Three different methods have been used to qualitatively assess the improvements of the camera parameters. The implementation of the proposed algorithm is publicly available at Furukawa and Ponce (2008b).

Carved visual hulls for image-based modeling

This article presents a novel method for acquiring high-quality solid models of complex 3D shapes from multiple calibrated photographs. After the purely geometric constraints associated with the silhouettes found in each image have been used to construct a coarse surface approximation in the form of a visual hull, photoconsistency constraints are enforced in three consecutive steps: (1) the rims where the surface grazes the visual hull are first identified through dynamic programming; (2) with the rims now fixed, the visual hull is carved using graph cuts to globally optimize the photoconsistency of the surface and recover its main features; (3) an iterative (local) refinement step is finally used to recover fine surface details. The proposed approach has been implemented, and experiments with six real data sets are presented, along with qualitative comparisons with several state-of-the-art image-based-modeling algorithms.

Dense 3D motion capture from synchronized video streams

This paper proposes a novel approach to non-rigid, markerless motion capture from synchronized video streams acquired by calibrated cameras. The instantaneous geometry of the observed scene is represented by a polyhedral mesh with fixed topology. The initial mesh is constructed in the first frame using the publicly available PMVS software for multi-view stereo [7]. Its deformation is captured by tracking its vertices over time, using two optimization processes at each frame: a local one using a rigid motion model in the neighborhood of each vertex, and a global one using a regularized nonrigid model for the whole mesh. Qualitative and quantitative experiments using seven real datasets show that our algorithm effectively handles complex nonrigid motions and severe occlusions.

Reconstructing the World's Museums

Virtual exploration tools for large indoor environments (e.g. museums) have so far been limited to either blueprint-style 2D maps that lack photo-realistic views of scenes, or ground-level image-to-image transitions, which are immersive but ill-suited for navigation. On the other hand, photorealistic aerial maps would be a useful navigational guide for large indoor environments, but it is impossible to directly acquire photographs covering a large indoor environment from aerial viewpoints. This paper presents a 3D reconstruction and visualization system for automatically producing clean and well-regularized texture-mapped 3D models for large indoor scenes, from ground-level photographs and 3D laser points. The key component is a new algorithm called “inverse constructive solid geometry (CSG)” for reconstructing a scene with a CSG representation consisting of volumetric primitives, which imposes powerful regularization constraints. We also propose several novel techniques to adjust the 3D model to make it suitable for rendering the 3D maps from aerial viewpoints. The visualization system enables users to easily browse a large-scale indoor environment from a bird’s-eye view, locate specific room interiors, fly into a place of interest, view immersive ground-level panorama views, and zoom out again, all with seamless 3D transitions. We demonstrate our system on various museums, including the Metropolitan Museum of Art in New York City—one of the largest art galleries in the world.

Projective Visual Hulls

This article presents a novel method for computing the visual hull of a solid bounded by a smooth surface and observed by a finite set of cameras. The visual hull is the intersection of the visual cones formed by back-projecting the silhouettes found in the corresponding images. We characterize its surface as a generalized polyhedron whose faces are visual cone patches; edges are intersection curves between two viewing cones; and vertices are frontier points where the intersection of two cones is singular, or intersection points where triples of cones meet. We use the mathematical framework of oriented projective differential geometry to develop an image-based algorithm for computing the visual hull. This algorithm works in a weakly calibrated setting–-that is, it only requires projective camera matrices or, equivalently, fundamental matrices for each pair of cameras. The promise of the proposed algorithm is demonstrated with experiments on several challenging data sets and a comparison to another state-of-the-art method.

Dense 3D motion capture for human faces

This paper proposes a novel approach to motion capture from multiple, synchronized video streams, specifically aimed at recording dense and accurate models of the structure and motion of highly deformable surfaces such as skin, that stretches, shrinks, and shears in the midst of normal facial expressions. Solving this problem is a key step toward effective performance capture for the entertainment industry, but progress so far has been hampered by the lack of appropriate local motion and smoothness models. The main technical contribution of this paper is a novel approach to regularization adapted to nonrigid tangential deformations. Concretely, we estimate the nonrigid deformation parameters at each vertex of a surface mesh, smooth them over a local neighborhood for robustness, and use them to regularize the tangential motion estimation. To demonstrate the power of the proposed approach, we have integrated it into our previous work for markerless motion capture [9], and compared the performances of the original and new algorithms on three extremely challenging face datasets that include highly nonrigid skin deformations, wrinkles, and quickly changing expressions. Additional experiments with a dataset featuring fast-moving cloth with complex and evolving fold structures demonstrate that the adaptability of the proposed regularization scheme to nonrigid tangential motion does not hamper its robustness, since it successfully recovers the shape and motion of the cloth without overfitting it despite the absence of stretch or shear in this case.