We present a benchmark for image-based 3D reconstruction. The benchmark sequences were acquired outside the lab, in realistic conditions. Ground-truth data was captured using an industrial laser scanner. The benchmark includes both outdoor scenes and indoor environments. High-resolution video sequences are provided as input, supporting the development of novel pipelines that take advantage of video input to increase reconstruction fidelity. We report the performance of many image-based 3D reconstruction pipelines on the new benchmark. The results point to exciting challenges and opportunities for future work.

Tanks and temples: benchmarking large-scale scene reconstruction

Motivated by the limitations of existing multi-view stereo benchmarks, we present a novel dataset for this task. Towards this goal, we recorded a variety of indoor and outdoor scenes using a high-precision laser scanner and captured both high-resolution DSLR imagery as well as synchronized low-resolution stereo videos with varying fields-of-view. To align the images with the laser scans, we propose a robust technique which minimizes photometric errors conditioned on the geometry. In contrast to previous datasets, our benchmark provides novel challenges and covers a diverse set of viewpoints and scene types, ranging from natural scenes to man-made indoor and outdoor environments. Furthermore, we provide data at significantly higher temporal and spatial resolution. Our benchmark is the first to cover the important use case of hand-held mobile devices while also providing high-resolution DSLR camera images. We make our datasets and an online evaluation server available at http://www.eth3d.net.

/pdf/a-multi-view-stereo-benchmark-with-high-resolution-images-3fxaoj7hp6.pdf

A Multi-view Stereo Benchmark with High-Resolution Images and Multi-camera Videos

We introduce a data-driven approach to complete partial 3D shapes through a combination of volumetric deep neural networks and 3D shape synthesis. From a partially-scanned input shape, our method first infers a low-resolution &#x2013; but complete &#x2013; output. To this end, we introduce a 3D-Encoder-Predictor Network (3D-EPN) which is composed of 3D convolutional layers. The network is trained to predict and fill in missing data, and operates on an implicit surface representation that encodes both known and unknown space. This allows us to predict global structure in unknown areas at high accuracy. We then correlate these intermediary results with 3D geometry from a shape database at test time. In a final pass, we propose a patch-based 3D shape synthesis method that imposes the 3D geometry from these retrieved shapes as constraints on the coarsely-completed mesh. This synthesis process enables us to reconstruct fine-scale detail and generate high-resolution output while respecting the global mesh structure obtained by the 3D-EPN. Although our 3D-EPN outperforms state-of-the-art completion method, the main contribution in our work lies in the combination of a data-driven shape predictor and analytic 3D shape synthesis. In our results, we show extensive evaluations on a newly-introduced shape completion benchmark for both real-world and synthetic data.

https://openaccess.thecvf.com/content_cvpr_2017/papers/Dai_Shape_Completion_Using_CVPR_2017_paper.pdf

Shape Completion Using 3D-Encoder-Predictor CNNs and Shape Synthesis

We introduce a data-driven approach to complete partial 3D shapes through a combination of volumetric deep neural networks and 3D shape synthesis. From a partially-scanned input shape, our method first infers a low-resolution -- but complete -- output. To this end, we introduce a 3D-Encoder-Predictor Network (3D-EPN) which is composed of 3D convolutional layers. The network is trained to predict and fill in missing data, and operates on an implicit surface representation that encodes both known and unknown space. This allows us to predict global structure in unknown areas at high accuracy. We then correlate these intermediary results with 3D geometry from a shape database at test time. In a final pass, we propose a patch-based 3D shape synthesis method that imposes the 3D geometry from these retrieved shapes as constraints on the coarsely-completed mesh. This synthesis process enables us to reconstruct fine-scale detail and generate high-resolution output while respecting the global mesh structure obtained by the 3D-EPN. Although our 3D-EPN outperforms state-of-the-art completion method, the main contribution in our work lies in the combination of a data-driven shape predictor and analytic 3D shape synthesis. In our results, we show extensive evaluations on a newly-introduced shape completion benchmark for both real-world and synthetic data.

Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesis

We propose the first fast and certifiable algorithm for the registration of two sets of three-dimensional (3-D) points in the presence of large amounts of outlier correspondences. A certifiable algorithm is one that attempts to solve an intractable optimization problem (e.g., robust estimation with outliers) and provides readily checkable conditions to verify if the returned solution is optimal (e.g., if the algorithm produced the most accurate estimate in the face of outliers) or bound its suboptimality or accuracy. Toward this goal, we first reformulate the registration problem using a truncated least squares (TLS) cost that makes the estimation insensitive to a large fraction of spurious correspondences. Then, we provide a general graph-theoretic framework to decouple scale, rotation, and translation estimation, which allows solving in cascade for the three transformations. Despite the fact that each subproblem (scale, rotation, and translation estimation) is still nonconvex and combinatorial in nature, we show that 1) TLS scale and (component-wise) translation estimation can be solved in polynomial time via an adaptive voting scheme, 2) TLS rotation estimation can be relaxed to a semidefinite program (SDP) and the relaxation is tight, even in the presence of extreme outlier rates, and 3) the graph-theoretic framework allows drastic pruning of outliers by finding the maximum clique. We name the resulting algorithm TEASER ( Truncated least squares Estimation And SEmidefinite Relaxation ). While solving large SDP relaxations is typically slow, we develop a second fast and certifiable algorithm, named TEASER++, that uses graduated nonconvexity to solve the rotation subproblem and leverages Douglas-Rachford Splitting to efficiently certify global optimality. For both algorithms, we provide theoretical bounds on the estimation errors, which are the first of their kind for robust registration problems. Moreover, we test their performance on standard benchmarks, object detection datasets, and the 3DMatch scan matching dataset, and show that 1) both algorithms dominate the state-of-the-art (e.g., RANSAC, branch-&-bound, heuristics) and are robust to more than $\text{99}\%$ outliers when the scale is known, 2) TEASER++ can run in milliseconds and it is currently the fastest robust registration algorithm, and 3) TEASER++ is so robust it can also solve problems without correspondences (e.g., hypothesizing all-to-all correspondences), where it largely outperforms ICP and it is more accurate than Go-ICP while being orders of magnitude faster. We release a fast open-source C++ implementation of TEASER++.

/pdf/teaser-fast-and-certifiable-point-cloud-registration-40gs80w55f.pdf

TEASER: Fast and Certifiable Point Cloud Registration

We present PatchmatchNet, a novel and learnable cascade formulation of Patchmatch for high-resolution multi-view stereo. With high computation speed and low memory requirement, PatchmatchNet can process higher resolution imagery and is more suited to run on resource limited devices than competitors that employ 3D cost volume regularization. For the first time we introduce an iterative multi-scale Patchmatch in an end-to-end trainable architecture and improve the Patchmatch core algorithm with a novel and learned adaptive propagation and evaluation scheme for each iteration. Extensive experiments show a very competitive performance and generalization for our method on DTU, Tanks & Temples and ETH3D, but at a significantly higher efficiency than all existing top-performing models: at least two and a half times faster than state-of-the-art methods with twice less memory usage.

PatchmatchNet: Learned Multi-View Patchmatch Stereo

In this paper, we propose an efficient and accurate scheme for the integration of multiple stereo-based depth measurements. For each provided depth map a confidence-based weight is assigned to each depth estimate by evaluating local geometry orientation, underlying camera setting and photometric evidence. Subsequently, all hypotheses are fused together into a compact and consistent 3D model. Thereby, visibility conflicts are identified and resolved, and fitting measurements are averaged with regard to their confidence scores. The individual stages of the proposed approach are validated by comparing it to two alternative techniques which rely on a conceptually different fusion scheme and a different confidence inference, respectively. Pursuing live 3D reconstruction on mobile devices as a primary goal, we demonstrate that the developed method can easily be integrated into a system for monocular interactive 3D modeling by substantially improving its accuracy while adding a negligible overhead to its performance and retaining its interactive potential.

/pdf/turning-mobile-phones-into-3d-scanners-31kj5vz5lv.pdf

Turning Mobile Phones into 3D Scanners

We present PatchmatchNet, a novel and learnable cascade formulation of Patchmatch for high-resolution multi-view stereo. With high computation speed and low memory requirement, PatchmatchNet can process higher resolution imagery and is more suited to run on resource limited devices than competitors that employ 3D cost volume regularization. For the first time we introduce an iterative multi-scale Patchmatch in an end-to-end trainable architecture and improve the Patchmatch core algorithm with a novel and learned adaptive propagation and evaluation scheme for each iteration. Extensive experiments show a very competitive performance and generalization for our method on DTU, Tanks & Temples and ETH3D, but at a significantly higher efficiency than all existing top-performing models: at least two and a half times faster than state-of-the-art methods with twice less memory usage. Code is available at https://github.com/FangjinhuaWang/PatchmatchNet.

/pdf/patchmatchnet-learned-multi-view-patchmatch-stereo-1l5w5g8u37.pdf

Image-based localization is a core component of many augmented/mixed reality (AR/MR) and autonomous robotic systems. Current localization systems rely on the persistent storage of 3D point clouds of the scene to enable camera pose estimation, but such data reveals potentially sensitive scene information. This gives rise to significant privacy risks, especially as for many applications 3D mapping is a background process that the user might not be fully aware of. We pose the following question: How can we avoid disclosing confidential information about the captured 3D scene, and yet allow reliable camera pose estimation? This paper proposes the first solution to what we call privacy preserving image-based localization. The key idea of our approach is to lift the map representation from a 3D point cloud to a 3D line cloud. This novel representation obfuscates the underlying scene geometry while providing sufficient geometric constraints to enable robust and accurate 6-DOF camera pose estimation. Extensive experiments on several datasets and localization scenarios underline the high practical relevance of our proposed approach.

/pdf/privacy-preserving-image-based-localization-4qvgfw25qq.pdf

Privacy Preserving Image-Based Localization

We propose a novel prior for variational 3D reconstruction that favors symmetric solutions when dealing with noisy or incomplete data. We detect symmetries from incomplete data while explicitly handling unexplored areas to allow for plausible scene completions. The set of detected symmetries is then enforced on their respective support domain within a variational reconstruction framework. This formulation also handles multiple symmetries sharing the same support. The proposed approach is able to denoise and complete surface geometry and even hallucinate large scene parts. We demonstrate in several experiments the benefit of harnessing symmetries when regularizing a surface.

/pdf/a-symmetry-prior-for-convex-variational-3d-reconstruction-387gkks2gn.pdf

Pablo Speciale

Papers

PatchmatchNet: Learned Multi-View Patchmatch Stereo

Turning Mobile Phones into 3D Scanners

PatchmatchNet: Learned Multi-View Patchmatch Stereo

Privacy Preserving Image-Based Localization

A Symmetry Prior for Convex Variational 3D Reconstruction