We present a novel real-time dense and semantic neural field mapping system that uses only monocular images as input. Our scene representation is a dense continuous radiance field represented by a Multi-Layer Perceptron (MLP), trained from scratch in real-time. We build on high-performance sparse visual SLAM and use camera poses and sparse keypoint depths as supervision alongside RGB keyframes. Since no prior training is required, our system flexibly fits to arbitrary scale and structure at runtime, and works even with strong specular reflections. We demonstrate reconstruction over a range of scenes from small indoor to large outdoor spaces. We also show that the method can straightforwardly benefit from additional inputs such as learned depth priors or semantic labels for more precise and advanced mapping.

There are many possibilities for how to represent the map in simultaneous localisation and mapping (SLAM). While sparse, keypoint-based SLAM systems have achieved impressive levels of accuracy and robustness, their maps may not be suitable for many robotic tasks. Dense SLAM systems are capable of producing dense reconstructions, but can be computationally expensive and, like sparse systems, lack higher-level information about the structure of a scene. Human-made environments contain a lot of structure, and we seek to take advantage of this by enabling the use of quadric surfaces as features in SLAM systems. We introduce a minimal representation for quadric surfaces and show how this can be included in a least-squares formulation. We also show how our representation can be easily extended to include additional constraints on quadrics such as those found in quadrics of revolution. Finally, we introduce a proof-of-concept SLAM system using our representation, and provide some experimental results using an RGB-D dataset.

We present a coarse-to-fine discretisation method that enables the use of discrete reinforcement learning approaches in place of unstable and data-inefficient actorcritic methods in continuous robotics domains. This approach builds on the recently released ARM algorithm, which replaces the continuous next-best pose agent with a discrete one, with coarse-to-fine Q-attention. Given a voxelised scene, coarse-to-fine Q-attention learns what part of the scene to ‘zoom’ into. When this ‘zooming’ behaviour is applied iteratively, it results in a near-lossless discretisation of the translation space, and allows the use of a discrete action, deep Q-learning method. We show that our new coarse-to-fine algorithm achieves state-of-the-art performance on several difficult sparsely rewarded RLBench vision-based robotics tasks, and can train real-world policies, tabula rasa, in a matter of minutes, with as little as 3 demonstrations.

By estimating 3D shape and instances from a single view, we can capture information about an environment quickly, without the need for comprehensive scanning and multi-view fusion. Solving this task for composite scenes (such as object stacks) is challenging: occluded areas are not only ambiguous in shape but also in instance segmentation; multiple decompositions could be valid. We observe that physics constrains decomposition as well as shape in occluded regions and hypothesise that a latent space learned from scenes built under physics simulation can serve as a prior to better predict shape and instances in occluded regions. To this end we propose SIMstack, a depth-conditioned Variational Auto-Encoder (VAE), trained on a dataset of objects stacked under physics simulation. We formulate instance segmentation as a centre voting task which allows for class-agnostic detection and doesn’t require setting the maximum number of objects in the scene. At test time, our model can generate 3D shape and instance segmentation from a single depth view, probabilistically sampling proposals for the occluded region from the learned latent space. Our method has practical applications in providing robots some of the ability humans have to make rapid intuitive inferences of partially observed scenes. We demonstrate an application for precise (non-disruptive) object grasping of unknown objects from a single depth view.

Semantic labelling is highly correlated with geometry and radiance reconstruction, as scene entities with similar shape and appearance are more likely to come from similar classes. Recent implicit neural reconstruction techniques are appealing as they do not require prior training data, but the same fully self-supervised approach is not possible for semantics because labels are human-defined properties.We extend neural radiance fields (NeRF) to jointly encode semantics with appearance and geometry, so that complete and accurate 2D semantic labels can be achieved using a small amount of in-place annotations specific to the scene. The intrinsic multi-view consistency and smoothness of NeRF benefit semantics by enabling sparse labels to efficiently propagate. We show the benefit of this approach when labels are either sparse or very noisy in room-scale scenes. We demonstrate its advantageous properties in various interesting applications such as an efficient scene labelling tool, novel semantic view synthesis, label denoising, super-resolution, label interpolation and multi-view semantic label fusion in visual semantic mapping systems.

The best way to combine the results of deep learning with standard 3D reconstruction pipelines remains an open problem. While systems that pass the output of traditional multi-view stereo approaches to a network for regularisation or refinement currently seem to get the best results, it may be preferable to treat deep neural networks as separate components whose results can be probabilistically fused into geometry- based systems. Unfortunately, the error models required to do this type of fusion are not well understood, with many different approaches being put forward. Recently, a few systems have achieved good results by having their networks predict probability distributions rather than single values. We propose using this approach to fuse a learned single-view depth prior into a standard 3D reconstruction system. Our system is capable of incrementally producing dense depth maps for a set of keyframes. We train a deep neural network to predict discrete, nonparametric probability distributions for the depth of each pixel from a single image. We then fuse this "probability volume" with another probability volume based on the photometric consistency between subsequent frames and the keyframe image. We argue that combining the probability volumes from these two sources will result in a volume that is better conditioned. To extract depth maps from the volume, we minimise a cost function that includes a regularisation term based on network predicted surface normals and occlusion boundaries. Through a series of experiments, we demonstrate that each of these components improves the overall performance of the system.

Estimating motion and surrounding geometry of a moving camera remains a challenging inference problem. From an information theoretic point of view, estimates should get better as more information is included, such as is done in dense SLAM, but this is strongly dependent on the validity of the underlying models. In the present paper, we use triangular meshes as both compact and dense geometry representation. To allow for simple and fast usage, we propose a view-based formulation for which we predict the in-plane vertex coordinates directly from images and then employ the remaining vertex depth components as free variables. Flexible and continuous integration of information is achieved through the use of a residual based inference technique. This so-called factor graph encodes all information as mapping from free variables to residuals, the squared sum of which is minimised during inference. We propose the use of different types of learnable residuals, which are trained end-to-end to increase their suitability as information bearing models and to enable accurate and reliable estimation. Detailed evaluation of all components is provided on both synthetic and real data which confirms the practicability of the presented approach.

While the keypoint-based maps created by sparse monocular Simultaneous Localisation and Mapping (SLAM) systems are useful for camera tracking, dense 3D reconstructions may be desired for many robotic tasks. Solutions involving depth cameras are limited in range and to indoor spaces, and dense reconstruction systems based on minimising the photometric error between frames are typically poorly constrained and suffer from scale ambiguity. To address these issues, we propose a 3D reconstruction system that leverages the output of a Convolutional Neural Network (CNN) to produce fully dense depth maps for keyframes that include metric scale. Our system, DeepFusion, is capable of producing real-time dense reconstructions on a GPU. It fuses the output of a semi-dense multiview stereo algorithm with the depth and gradient predictions of a CNN in a probabilistic fashion, using learned uncertainties produced by the network. While the network only needs to be run once per keyframe, we are able to optimise for the depth map with each new frame so as to constantly make use of new geometric constraints. Based on its performance on synthetic and real world datasets, we demonstrate that DeepFusion is capable of performing at least as well as other comparable systems.

While dense visual SLAM methods are capable of estimating dense reconstructions of the environment, they suffer from a lack of robustness in their tracking step, especially when the optimisation is poorly initialised. Sparse visual SLAM systems have attained high levels of accuracy and robustness through the inclusion of inertial measurements in a tightly-coupled fusion. Inspired by this performance, we propose the first tightly-coupled dense RGB-D-inertial SLAM system. Our system has real-time capability while running on a GPU. It jointly optimises for the camera pose, velocity, IMU biases and gravity direction while building up a globally consistent, fully dense surfel-based 3D reconstruction of the environment. Through a series of experiments on both synthetic and real world datasets, we show that our dense visual-inertial SLAM system is more robust to fast motions and periods of low texture and low geometric variation than a related RGB-D-only SLAM system.