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.
Visual localization and mapping is the key technology underlying the majority of Mixed Reality and robotics systems. Most state-of-the-art approaches rely on local features to establish correspondences between images. In this paper, we present three novel scenarios for localization and mapping which require the continuous update of feature representations and the ability to match across different feature types. While localization and mapping is a fundamental computer vision problem, the traditional setup treats it as a single-shot process using the same local image features throughout the evolution of a map. This assumes the whole process is repeated from scratch whenever the underlying features are changed. However, reiterating it is typically impossible in practice, because raw images are often not stored and re-building the maps could lead to loss of the attached digital content. To overcome the limitations of current approaches, we present the first principled solution to cross-descriptor localization and mapping. Our data-driven approach is agnostic to the feature descriptor type, has low computational requirements, and scales linearly with the number of description algorithms. Extensive experiments demonstrate the effectiveness of our approach on state-of-the-art benchmarks for a variety of handcrafted and learned features.
Objects moving at high speed appear significantly blurred when captured with cameras. The blurry appearance is especially ambiguous when the object has complex shape or texture. In such cases, classical methods, or even humans, are unable to recover the object's appearance and motion. We propose a method that, given a single image with its estimated background, outputs the object's appearance and position in a series of sub-frames as if captured by a high-speed camera (i.e. temporal super-resolution). The proposed generative model embeds an image of the blurred object into a latent space representation, disentangles the background, and renders the sharp appearance. Inspired by the image formation model, we design novel self-supervised loss function terms that boost performance and show good generalization capabilities. The proposed DeFMO method is trained on a complex synthetic dataset, yet it performs well on real-world data from several datasets. DeFMO outperforms the state of the art and generates high-quality temporal super-resolution frames.
We present a novel online depth map fusion approach that learns depth map aggregation in a latent feature space. While previous fusion methods use an explicit scene representation like signed distance functions (SDFs), we propose a learned feature representation for the fusion. The key idea is a separation between the scene representation used for the fusion and the output scene representation, via an additional translator network. Our neural network architecture consists of two main parts: a depth and feature fusion sub-network, which is followed by a translator sub-network to produce the final surface representation (e.g. TSDF) for visualization or other tasks. Our approach is real-time capable, handles high noise levels, and is particularly able to deal with gross outliers common for photometric stereo-based depth maps. Experiments on real and synthetic data demonstrate improved results compared to the state of the art, especially in challenging scenarios with large amounts of noise and outliers.
Localization of a robotic system within a previously mapped environment is important for reducing estimation drift and for reusing previously built maps. Existing techniques for geometry-based localization have focused on the description of local surface geometry, usually using pointclouds as the underlying representation. We propose a system for geometry-based localization that extracts features directly from an implicit surface representation: the Signed Distance Function (SDF). The SDF varies continuously through space, which allows the proposed system to extract and utilize features describing both surfaces and free-space. Through evaluations on public datasets, we demonstrate the flexibility of this approach, and show an increase in localization performance over state-of-the-art handcrafted surfaces-only descriptors. We achieve an average improvement of ~12% on an RGB-D dataset and ~18% on a LiDAR-based dataset. Finally, we demonstrate our system for localizing a LiDAR-equipped MAV within a previously built map of a search and rescue training ground.
Most of the current scene flow methods choose to model scene flow as a per point translation vector without differentiating between static and dynamic components of 3D motion. In this work we present an alternative method for end-to-end scene flow learning by joint estimation of non-rigid residual flow and ego-motion flow for dynamic 3D scenes. We propose to learn the relative rigid transformation from a pair of point clouds followed by an iterative refinement. We then learn the non-rigid flow from transformed inputs with the deducted rigid part of the flow. Furthermore, we extend the supervised framework with self-supervisory signals based on the temporal consistency property of a point cloud sequence. Our solution allows both training in a supervised mode complemented by self-supervisory loss terms as well as training in a fully self-supervised mode. We demonstrate that decomposition of scene flow into non-rigid flow and ego-motion flow along with an introduction of the self-supervisory signals allowed us to outperform the current state-of-the-art supervised methods.
Localization of a robotic system within a previously mapped environment is important for reducing estimation drift and for reusing previously built maps. Existing techniques for geometry-based localization have focused on the description of local surface geometry, usually using pointclouds as the underlying representation. We propose a system for geometry-based localization that extracts features directly from an implicit surface representation: the Signed Distance Function (SDF). The SDF varies continuously through space, which allows the proposed system to extract and utilize features describing both surfaces and free-space. Through evaluations on public datasets, we demonstrate the flexibility of this approach, and show an increase in localization performance over state-of-the-art handcrafted surfaces-only descriptors. We achieve an average improvement of ~12% on an RGB-D dataset and ~18% on a LiDAR-based dataset. Finally, we demonstrate our system for localizing a LiDAR-equipped MAV within a previously built map of a search and rescue training ground.
Representing scenes at the granularity of objects is a prerequisite for scene understanding and decision making. We propose a novel approach for learning multi-object 3D scene representations from images. A recurrent encoder regresses a latent representation of 3D shapes, poses and texture of each object from an input RGB image. The 3D shapes are represented continuously in function-space as signed distance functions (SDF) which we efficiently pre-train from example shapes in a supervised way. By differentiable rendering we then train our model to decompose scenes self-supervised from RGB-D images. Our approach learns to decompose images into the constituent objects of the scene and to infer their shape, pose and texture from a single view. We evaluate the accuracy of our model in inferring the 3D scene layout and demonstrate its generative capabilities.
Surface reconstruction from magnetic resonance (MR) imaging data is indispensable in medical image analysis and clinical research. A reliable and effective reconstruction tool should: be fast in prediction of accurate well localised and high resolution models, evaluate prediction uncertainty, work with as little input data as possible. Current deep learning state of the art (SOTA) 3D reconstruction methods, however, often only produce shapes of limited variability positioned in a canonical position or lack uncertainty evaluation. In this paper, we present a novel probabilistic deep learning approach for concurrent 3D surface reconstruction from sparse 2D MR image data and aleatoric uncertainty prediction. Our method is capable of reconstructing large surface meshes from three quasi-orthogonal MR imaging slices from limited training sets whilst modelling the location of each mesh vertex through a Gaussian distribution. Prior shape information is encoded using a built-in linear principal component analysis (PCA) model. Extensive experiments on cardiac MR data show that our probabilistic approach successfully assesses prediction uncertainty while at the same time qualitatively and quantitatively outperforms SOTA methods in shape prediction. Compared to SOTA, we are capable of properly localising and orientating the prediction via the use of a spatially aware neural network.