Image-based localization in GNSS-denied environments is critical for UAV autonomy. Existing state-of-the-art approaches rely on matching UAV images to geo-referenced satellite images; however, they typically require large-scale, paired UAV–satellite datasets for training. Such data are costly to acquire and often unavailable, limiting their applicability. To address this challenge, we adopt a training paradigm that removes the need for UAV imagery during training by learning directly from satellite-view reference images. This is achieved through a dedicated augmentation strategy that simulates the visual domain shift between satellite and real-world UAV views. We introduce CAEVL, an efficient model designed to exploit this paradigm, and validate it on ViLD, a new and challenging dataset of real-world UAV images that we release to the community. Our method achieves competitive performance compared to approaches trained with paired data, demonstrating its effectiveness and strong generalization capabilities.

We propose a new method for the geolocalization of Un-maned Aerial Vehicles (UAV) in environments without Global Navigation Stallite Systems (GNSS). Current state-of-the-art methods use an offline-trained encoder to compute a vector representation (embedding) of the current UAV’s view, and compare it with the pre-computed embeddings of geo-referenced images in order to deduce the UAV’s position. Here, we show that the performance of these methods can be greatly improved by pre-processing the images by extracting their edges, which are robust to seasonal and illumination changes. Moreover, we also show that using edges improves the robustness to orientation and altitude errors. Finally, we present a confidence criterion for localization. Our findings are validated using synthetic experiments.

Neural implicit surfaces have become an important technique for multi-view 3D reconstruction but their accuracy remains limited. In this paper, we argue that this comes from the difficulty to learn and render high frequency textures with neural networks. We thus propose to add to the standard neural rendering optimization a direct photo-consistency term across the different views. Intuitively, we optimize the implicit geometry so that it warps views on each other in a consistent way. We demonstrate that two elements are key to the success of such an approach: (i) warping entire patches, using the predicted occupancy and normals of the 3D points along each ray, and measuring their similarity with a robust structural similarity (SSIM); (ii) handling visibility and occlusion in such a way that incorrect warps are not given too much importance while encouraging a reconstruction as complete as possible. We evaluate our approach, dubbed NeuralWarp, on the standard DTU and EPFL benchmarks and show it outperforms state of the art unsupervised implicit surfaces reconstructions by over 20% on both datasets. Our code is available at https://github.com/fdarmon/NeuralWarp

Deep multi-view stereo (MVS) methods have been developed and extensively compared on simple datasets, where they now outperform classical approaches. In this paper, we ask whether the conclusions reached in controlled scenarios are still valid when working with Internet photo collections. We propose a methodology for evaluation and explore the influence of three aspects of deep MVS methods: network architecture, training data, and supervision. We make several key observations, which we extensively validate quantitatively and qualitatively, both for depth prediction and complete 3D reconstructions. First, complex unsupervised approaches cannot train on data in the wild. Our new approach makes it possible with three key elements: upsampling the output, softmin based aggregation and a single reconstruction loss. Second, supervised deep depthmap-based MVS methods are state-of-the art for reconstruction of few internet images. Finally, our evaluation provides very different results than usual ones. This shows that evaluation in uncontrolled scenarios is important for new architectures.

Describes our 3D graphical interface for industrial process control. The software is based on Visio and WinCC. The software first builds a 3D model of the factory. This can be done using user input or automatically from industrial drawings (electronic or paper copies). Secondly, process values are attached to the scene and the process is activated and controlled through WinCC.

This paper presents a new approach to image-based guidance of a needle or surgical tool during percutaneous procedures. The method is based on visual servoing. It requires no prior calibration or registration. The technique provides highly precise 3D-alignment of the tool with respect to an anatomic target. By taking advantage of projective geometry and projective invariants, this can be achieved in a fixed number (12) of iterations. In addition the approach estimates the required insertion depth. Experiments include automatic 3D alignment and insertion of a needle held by a medical robot into a pig kidney under X-ray fluoroscopy.

Image segmentation has traditionally been thought of us a low/mid-level vision process incorporating no high level constraints. However, in complex and uncontrolled environments, such bottom-up strategies have drawbacks that lead to large misclassification rates. Remedies to this situation include taking into account (1) contextual and application constraints, (2) user input and feedback to incrementally improve the performance of the system. We attempt to incorporate these in the context of pipeline segmentation in industrial images. This problem is of practical importance for the 3D reconstruction of factory environments. However it poses several fundamental challenges mainly due to shading. Highlights and textural variations, etc. Our system performs pipe segmentation by fusing methods from physics-based vision, edge and texture analysis, probabilistic learning and the use of the graph-cut formalism.

Some issues in markerless tracking of human body motion are addressed. Extended Kalman filters have commonly been applied to kinematic variables, to combine predictions consistent with plausible motion, with the incoming stream of visual measurements. Kalman filtering is applicable only when the underlying distribution is approximately Gaussian. Often this assumption proves remarkably robust. There are two pervasive circumstances under which the Gaussianity assumption can break down. The first is kinematic singularity and the second is at joint endstops. Failure of Kalman filtering under these circumstances is illustrated. The non-Gaussian nature of the distributions is demonstrated experimentally by means of Monte Carlo simulation. Random simulation (particle filtering or Condensation) proves to provide a robust alternative algorithm for tracking that can also deal with these difficult conditions.

We explore the application of facial tracking to automated re-animation. To this end, it is necessary to recover both head-pose and facial expression from the facial movement of a performer. However, both effects are coupled. This is a serious problem, which previous studies haven't fully considered. The solution to this interaction problem proposed here is to solve explicitly, at each timestep, for pose and expression variables. In principle this is a nonlinear inverse problem. However, appropriate parameterisation of pose in terms of affine transformations with parallax, and of expression in terms of key-frames, reduces the problem to a bilinear one. This can then be solved directly by Singular Value Decomposition. Thus actor-driven animation has ben implemented in real-time, at video field-rate, using two Indy desktop workstations.

The paper describes an approach to the tracking of complex shapes through image sequences, that combines deformable region models and deformable contours. A deformable region model is presented: its optimisation is based on texture correlation and is constrained by the use of a motion model, such as rigid, affine or homographic. The use of texture information (versus edge information) noticeably improves the tracking performances of deformable models in the presence of texture. Then the region contour is refined using an edge based deformable model in order to better deal with specularities, non planar objects and occlusions. The method is illustrated and validated by experimental results on real images.>

This paper describes a new approach to track complex primitives along image sequences - integrating snake-based contour tracking and region-based motion analysis. First, a snake tracks the region outline and performs segmentation. Then the motion of the extracted region is estimated by a dense analysis of the apparent motion over the region, using spatio-temporal image gradients. Finally, this motion measurement is filtered to predict the region location in the next frame, and thus to guide (i.e. to initialize) the tracking snake in the next frame. Therefore, these two approaches collaborate and exchange information to overcome the limitations of each of them. The method is illustrated by experimental results on real images.

The authors propose to replace a set of connected 3-D points or line segments with a 3-D formable curve, using weighted least-squares approximation. The parameters of the 3-D energy-minimizing curve are updated so that its 2-D projections on two or three stereoscopic images converge toward the corresponding image edges. Thus, a more realistic representation of the real non-polygonal world is obtained. The case where no initial 3-D information is provided is also considered. A novel approach is proposed to match a given 2-D curve in the first image to its corresponding position in the second image. The curve is tracked between the two images by a deformable curve constrained by a 2-D motion model. The position of the 2-D curve in the second image is then refined by relaxing the motion constraint. As a result, the correspondences between the curves of both images are known. An initial estimate of the 3-D curve can be recovered using the calibration parameters of the stereoscopic system. These approaches are illustrated by experimental results.>

Deals with the development and implementation of several deformable models (snakes) designed to extract some particular and important features in an image. First, the authors propose polygonal-like snakes which detect corners and triple junctions in an image. In the second part, they describe parametric snakes designed in order to extract curves of degree greater than one: segments, ellipses, superellipses and rational B-spline curves.>