This paper presents Planar Gaussian Splatting (PGS), a novel neural rendering approach to learn the 3D geometry and parse the 3D planes of a scene, directly from multiple RGB images. The PGS leverages Gaussian primitives to model the scene and employ a hierarchical Gaussian mixture approach to group them. Similar Gaussians are progressively merged probabilistically in the tree-structured Gaussian mixtures to identify distinct 3D plane instances and form the overall 3D scene geometry. In order to enable the grouping, the Gaussian primitives contain additional parameters, such as plane descriptors derived by lifting 2D masks from a general 2D segmentation model and surface normals. Experiments show that the proposed PGS achieves state-of-the-art performance in 3D planar reconstruction without requiring either 3D plane labels or depth supervision. In contrast to existing supervised methods that have limited generalizability and struggle under domain shift, PGS maintains its performance across datasets thanks to its neural rendering and scene-specific optimization mechanism, while also being significantly faster than existing optimization-based approaches.

Humans perceive and construct the surrounding world as an arrangement of simple parametric models. In particular, man-made environments commonly consist of volumetric primitives such as cuboids or cylinders. Inferring these primitives is an important step to attain high-level, abstract scene descriptions. Previous approaches directly estimate shape parameters from a 2D or 3D input, and are only able to reproduce simple objects, yet unable to accurately parse more complex 3D scenes. In contrast, we propose a robust estimator for primitive fitting, which can meaningfully abstract real-world environments using cuboids. A RANSAC estimator guided by a neural network fits these primitives to 3D features, such as a depth map. We condition the network on previously detected parts of the scene, thus parsing it one-by-one. To obtain 3D features from a single RGB image, we additionally optimise a feature extraction CNN in an end-to-end manner. However, naively minimising point-to-primitive distances leads to large or spurious cuboids occluding parts of the scene behind. We thus propose an occlusion-aware distance metric correctly handling opaque scenes. The proposed algorithm does not require labour-intensive labels, such as cuboid annotations, for training. Results on the challenging NYU Depth v2 dataset demonstrate that the proposed algorithm successfully abstracts cluttered real-world 3D scene layouts.

Dynamic scene graph generation aims at generating a scene graph of the given video. Compared to the task of scene graph generation from images, it is more challenging because of the dynamic relationships between objects and the temporal dependencies between frames allowing for a richer semantic interpretation. In this paper, we propose Spatial-temporal Transformer (STTran), a neural network that consists of two core modules: (1) a spatial encoder that takes an input frame to extract spatial context and reason about the visual relationships within a frame, and (2) a temporal decoder which takes the output of the spatial encoder as input in order to capture the temporal dependencies between frames and infer the dynamic relationships. Furthermore, STTran is flexible to take varying lengths of videos as input without clipping, which is especially important for long videos. Our method is validated on the benchmark dataset Action Genome (AG). The experimental results demonstrate the superior performance of our method in terms of dynamic scene graphs. Moreover, a set of ablative studies is conducted and the effect of each proposed module is justified. Code available at: https://github.com/yrcong/STTran.

This work presents a data-driven approach for the indoor localization of an observer on a 2D topological map of the environment. State-of-the-art techniques may yield accurate estimates only when they are tailor-made for a specific data modality like camera-based system that prevents their applicability to broader domains. Here, we establish a modality-agnostic framework (called OT-Isomap) and formulate the localization problem in the context of parametric manifold learning while leveraging optimal transportation. This framework allows jointly learning a low-dimensional embedding as well as correspondences with a topological map. We examine the generalizability of the proposed algorithm by applying it to data from diverse modalities such as image sequences and radio frequency signals. The experimental results demonstrate decimeter-level accuracy for localization using different sensory inputs.

In this work, we present a new versatile 3D multilinear statistical face model, based on a tensor factorisation of 3D face scans, that decomposes the shapes into person and expression subspaces. Investigation of the expression subspace reveals an inherent low-dimensional substructure, and further, a star-shaped structure. This is due to two novel findings. (1) Increasing the strength of one emotion approximately forms a linear trajectory in the subspace. (2) All these trajectories intersect at a single point - not at the neutral expression as assumed by almost all prior works-but at an apathetic expression. We utilise these structural findings by reparameterising the expression subspace by the fourth-order moment tensor centred at the point of apathy. We propose a 3D face reconstruction method from single or multiple 2D projections by assuming an uncalibrated projective camera model. The non-linearity caused by the perspective projection can be neatly included into the model. The proposed algorithm separates person and expression subspaces convincingly, and enables flexible, natural modelling of expressions for a wide variety of human faces. Applying the method on independent faces showed that morphing between different persons and expressions can be performed without strong deformations.

We present a robust estimator for fitting multiple parametric models of the same form to noisy measurements. Applications include finding multiple vanishing points in man-made scenes, fitting planes to architectural imagery, or estimating multiple rigid motions within the same sequence. In contrast to previous works, which resorted to hand-crafted search strategies for multiple model detection, we learn the search strategy from data. A neural network conditioned on previously detected models guides a RANSAC estimator to different subsets of all measurements, thereby finding model instances one after another. We train our method supervised, as well as, self-supervised. For supervised training of the search strategy, we contribute a new dataset for vanishing point estimation. Leveraging this dataset, the proposed algorithm is superior with respect to other robust estimators, as well as, to designated vanishing point estimation algorithms. For self-supervised learning of the search, we evaluate the proposed algorithm on multi-homography estimation and demonstrate an accuracy that is superior to state-of-the-art methods.

The horizon line is an important geometric feature for many image processing and scene understanding tasks in computer vision. For instance, in navigation of autonomous vehicles or driver assistance, it can be used to improve 3D reconstruction as well as for semantic interpretation of dynamic environments. While both algorithms and datasets exist for single images, the problem of horizon line estimation from video sequences has not gained attention. In this paper, we show how convolutional neural networks are able to utilise the temporal consistency imposed by video sequences in order to increase the accuracy and reduce the variance of horizon line estimates. A novel CNN architecture with an improved residual convolutional LSTM is presented for temporally consistent horizon line estimation. We propose an adaptive loss function that ensures stable training as well as accurate results. Furthermore, we introduce an extension of the KITTI dataset which contains precise horizon line labels for 43699 images across 72 video sequences. A comprehensive evaluation shows that the proposed approach consistently achieves superior performance compared with existing methods.

Roads constitute a major part of the lives of everybody. Heavy use, for instance by cars and especially trucks, and even soil movement lead to visible damages. While major roads are regularly inspected, smaller roads often lack attention. It is therefore of great interest to have camera-based systems which can automatically detect and even classify damages.This report presents a system developed by the authors as part of the Road Damage Detection and Classification Challenge at the 2018 IEEE Big Data Cup [1]. Further contributions made here are techniques to augment the small set of training data. As a major contribution we also propose refinements to the dataset and evaluation metric to improve the challenge.

In recent years, data-driven methods have shown great success for extracting information about the infrastructure in urban areas. These algorithms are usually trained on large datasets consisting of thousands or millions of labeled training examples. While large datasets have been published regarding cars, for cyclists very few labeled data is available although appearance, point of view, and positioning of even relevant objects differ. Unfortunately, labeling data is costly and requires a huge amount of work. In this paper, we thus address the problem of learning with very few labels. The aim is to recognize particular traffic signs in crowdsourced data to collect information which is of interest to cyclists. We propose a system for object recognition that is trained with only 15 examples per class on average. To achieve this, we combine the advantages of convolutional neural networks and random forests to learn a patch-wise classifier. In the next step, we map the random forest to a neural network and transform the classifier to a fully convolutional network. Thereby, the processing of full images is significantly accelerated and bounding boxes can be predicted. Finally, we integrate data of the Global Positioning System (GPS) to localize the predictions on the map. In comparison to Faster R-CNN and other networks for object recognition or algorithms for transfer learning, we considerably reduce the required amount of labeled data. We demonstrate good performance on the recognition of traffic signs for cyclists as well as their localization in maps.

This paper aims at simplified high dynamic range (HDR) image generation with non-modified, conventional camera sensors. One typical HDR approach is exposure bracketing, e.g. with varying shutter speeds. It requires to capture the same scene multiple times at different exposure times. These pictures are then merged into a single HDR picture which typically is converted back to an 8-bit image by using tone-mapping. Existing works on HDR imaging focus on image merging and tone mapping whereas we aim at simplified image acquisition. The proposed algorithm can be used in consumer-level cameras without hardware modifications at sensor level. Based on intermediate samplings of each sensor element during the total (pre-defined) exposure time, we extrapolate the luminance of sensor elements which are saturated after the total exposure time. Compared to existing HDR approaches which typically require three different images with carefully determined exposure times, we only take one image at the longest exposure time. The shortened total time between start and end of image acquisition can reduce ghosting artifacts. The experimental evaluation demonstrates the effectiveness of the algorithm.

In this paper, we present a new statistical model for human faces. Our approach is built upon a tensor factorisation model that allows controlled estimation, morphing and transfer of new facial shapes and expressions. We propose a direct parametrisation and regularisation for person and expression related terms so that the training database is well utilised. In contrast to existing works we are the first to reveal that the expression subspace is star shaped. This stems from the fact that increasing the strength of an expression approximately forms a linear trajectory in the expression subspace, and all these linear trajectories intersect in a single point which corresponds to the point of no expression or the point of apathy. After centring our analysis to this point, we then demonstrate how the dimensionality of the expression subspace can be further reduced by projection pursuit with the help of the fourth-order moment tensor. The results show that our method is able to achieve convincing separation of the person specific and expression subspaces as well as flexible, natural modelling of facial expressions for wide variety of human faces. By the proposed approach, one can morph between different persons and different expressions even if they do not exist in the database. In contrast to the state-of-the-art, the morphing works without causing strong deformations. In the application of expression classification, the results are also better.

For the generation of overview panoramic images from aerial surveillance videos, registered video frames are stitched together. Assuming a planar landscape, feature points can be detected and used to estimate a homography. However, if the features are affected by radial distortion, their mapping depends on their position within the frame and the resulting homography becomes inaccurate. As a result, the length of aerial panorama images is typically restricted to several hundred frames. To overcome this issue, we derive a model for the joint estimation of several homographies and one constant radial distortion. Due to the computational complexity of the solution, we propose a fast, iterative algorithm. Based on geometrical constraints, we regularize the projection of a jointly estimated picture group. We present panorama images from uncalibrated aerial videos with more than 1500 frames.

This article tackles the problem of estimating non-rigid human 3D shape and motion from image sequences taken by uncalibrated cameras. Similar to other state-of-the-art solutions we factorize 2D observations in camera parameters, base poses and mixing coefficients. Existing methods require sufficient camera motion during the sequence to achieve a correct 3D reconstruction. To obtain convincing 3D reconstructions from arbitrary camera motion, our method is based on a-priorly trained base poses. We show that strong periodic assumptions on the coefficients can be used to define an efficient and accurate algorithm for estimating periodic motion such as walking patterns. For the extension to non-periodic motion we propose a novel regularization term based on temporal bone length constancy. In contrast to other works, the proposed method does not use a predefined skeleton or anthropometric constraints and can handle arbitrary camera motion. We achieve convincing 3D reconstructions, even under the influence of noise and occlusions. Multiple experiments based on a 3D error metric demonstrate the stability of the proposed method. Compared to other state-of-the-art methods our algorithm shows a significant improvement.

Social network analysis is leveraged in a variety of applications such as identifying influential entities, detecting communities with special interests, and determining the flow of information and innovations. However, existing approaches for extracting social networks from unstructured Web content do not scale well and are only feasible for small graphs. In this paper, we introduce novel methodologies for query-based search engine mining, enabling efficient extraction of social networks from large amounts of Web data. To this end, we use patterns in phrase queries for retrieving entity connections, and employ a bootstrapping approach for iteratively expanding the pattern set. Our experimental evaluation in different domains demonstrates that our algorithms provide high quality results and allow for scalable and efficient construction of social graphs.

In this paper we propose a multilinear model of human pose and body shape which is estimated from a database of registered 3D body scans in different poses. The model is generated by factorizing the measurements into pose and shape dependent components. By combining it with an ICP based registration method, we are able to estimate pose and body shape of dressed subjects from single images. If several images of the subject are available, shape and poses can be optimized simultaneously for all input images. Additionally, while estimating pose and shape, we use the model as a virtual calibration pattern and also recover the parameters of the perspective camera model the images were created with.

The problem of reconstructing a 3D scene from a moving camera can be solved by means of the so-called Factorization method. It directly computes a global solution without the need to merge several partial reconstructions. However, if the trajectories are not complete, i.e. not every feature point could be observed in all the images, this method cannot be used. We use a Factorization-style algorithm for recovering the unobserved feature positions in a non-incremental way. This method uniformly utilizes all data and finds a global solution without any need of sequential or hierarchical merging. Two contributions are made in this work: Firstly, partially known trajectories are completed by minimizing the distance between the subspace and the trajectory within an affine subspace associated with the trajectory. This amounts to imposing a global constraint on the data. Secondly, we propose to further include local constraints derived from epipolar geometry into the estimation. It is shown how to simultaneously optimize both constraints. By using simulated and real image sequences we show the improvements achieved with our algorithm.