Abstract Video inpainting is the task of filling a region in a video in a visually convincing manner It is very challenging due to the high dimensionality of the data and the temporal consistency required for obtaining convincing results. Recently, diffusion models have shown impressive results in modeling complex data distributions, including images and videos. Such models remain nonetheless very expensive to train and to perform inference with, which strongly reduce their applicability to videos, and yields unreasonable computational loads. We show that in the case of video inpainting, thanks to the highly auto‐similar nature of videos, the training data of a diffusion model can be restricted to the input video and still produce very satisfying results. With this internal learning approach, where the training data is limited to a single video, our lightweight models perform very well with only half a million parameters, in contrast to the very large networks with billions of parameters typically found in the literature. We also introduce a new method for efficient training and inference of diffusion models in the context of internal learning, by splitting the diffusion process into different learning intervals corresponding to different noise levels of the diffusion process. We show qualitative and quantitative results, demonstrating that our method reaches or exceeds state of the art performance in the case of dynamic textures and complex dynamic backgrounds.

Abstract Film grain refers to the specific texture of film‐acquired images, due to the physical nature of photographic film. Being a visual signature of such images, there is a strong interest in the film‐industry for the rendering of these textures for digital images. Some previous works are able to closely mimic the physics of films and produce high quality results, but are computationally expensive. We propose a method based on a lightweight neural network and a texture aware loss function, achieving realistic results with very low complexity, even for large grains and high resolutions. We evaluate our algorithm both quantitatively and qualitatively with respect to previous work.

State-of-the-art red-green–blue (RGB) texture synthesis algorithms rely on style distances that are computed through statistics of deep features. These deep features are extracted by classification neural networks that have been trained on large datasets of RGB images. Extending such synthesis methods to multispectral images is not straightforward, since the pretrained networks are designed for and have been trained on RGB images. In this work, we propose two solutions to extend these methods to multispectral imaging (MSI). Neither of them requires additional training of the neural network from which the second-order neural statistics are extracted. The first one involves optimizing over batches of random triplets of spectral bands during training. The second one projects multispectral pixels onto a 3-D space. We further explore the benefit of a color transfer operation upstream of the projection to avoid the potentially abnormal color distributions induced by the projection. Our experiments compare the performances of the various methods through different metrics. We demonstrate that they can be used to perform exemplar-based texture synthesis, achieve good visual quality, and come close to state-of-the-art methods on RGB bands. Code is available athttps://github.com/selim2483/multispectraltextureCNN.

Deep learning methods for image restoration have produced impressive results over recent years. Nevertheless, they generalize poorly and need large learning image datasets to be collected for each new acquisition modality. In order to avoid the building of such datasets, it has been recently proposed to develop synthetic image datasets for training image restoration methods, using scale invariant dead leaves models. While the geometry of such models can be successfully encoded with only a few parameters, the color content cannot be straightforwardly encoded. In this paper, we leverage the concept of color lines prior to build a light parametric color model relying on a chromaticity/luminance factorization. Further, we show that the corresponding synthetic dataset can be used to train neural networks for the denoising of RAW images from different camera-phones, without using any image from these devices. This shows the potential of our approach to increase the generalization capacity of learning-based denoising approaches in real case scenarios.

Image generation is the task of producing new samples from one or several example images. Until recently, this has been done using large image databases, in particular using Generative Adversarial Networks (GANs). However, Shaham et al. [1] recently proposed the SinGAN method, which achieves this generation using a single image example. At the same time, researchers are realizing that classical patch- based methods can replace certain neural networks, with no costly training. In this paper, we present a purely patch-based method, named Patches for Single image generation (PSin), which requires no training and generates samples in seconds. Our algorithm is based on the minimization of a global, patch- based energy functional, which ensures the visual fidelity of the result to the original image. We also ensure diversity of the results by carefully choosing the initialization of the algorithm. We propose two initialization variants. We compare our results to both the original SinGAN and another recent patch-based image generation approach, both qualitatively and quantitatively using multiple metrics.

Deep learning based methods for single-image super-resolution (SR) have drawn a lot of attention lately. In particular, various papers have shown that the learning stage can be performed on a single image, resulting in the so-called internal approaches. The SinGAN method is one of these contributions, where the distribution of image patches is learnt on the image at hand and propagated at finer scales. Now, there are situations where some statistical a priori can be assumed for the final image. In particular, many natural phenomena yield images having power law Fourier spectrum, such as clouds and other texture images. In this work, we show how such a priori information can be integrated into an internal super-resolution approach, by constraining the learned up-sampling procedure of SinGAN. We consider various types of constraints, related to the Fourier power spectrum, the color histograms and the consistency of the upsampling scheme. We demonstrate on various experiments that these constraints are indeed satisfied, but also that some perceptual quality measures can be improved by the proposed approach.

High quality facial image editing is a challenging problem in the movie post-production industry, requiring a high degree of control and identity preservation. Previous works that attempt to tackle this problem may suffer from the entanglement of facial attributes and the loss of the person’s identity. Furthermore, many algorithms are limited to a certain task. To tackle these limitations, we propose to edit facial attributes via the latent space of a StyleGAN generator, by training a dedicated latent transformation network and incorporating explicit disentanglement and identity preservation terms in the loss function. We further introduce a pipeline to generalize our face editing to videos. Our model achieves a disentangled, controllable, and identity-preserving facial attribute editing, even in the challenging case of real (i.e., non-synthetic) images and videos. We conduct extensive experiments on image and video datasets and show that our model outperforms other state-of-the-art methods in visual quality and quantitative evaluation. Source codes are available at https://github.com/InterDigitalInc/latent-transformer.

Recent work has demonstrated the great potential of image editing in the latent space of powerful deep generative models such as StyleGAN. However, the success of such methods relies on the assumption that a linear hyperplane may separate the latent space into two subspaces for a binary attribute. In this work, we show that this hypothesis is a significant limitation and propose to learn a non-linear, regularized and identity-preserving latent space transformation that leads to more accurate and disentangled manipulations of facial attributes.

Face age editing has become a crucial task in film post-production, and is also becoming popular for general purpose photography. Recently, adversarial training has produced some of the most visually impressive results for image manipulation, including the face aging/de-aging task. In spite of considerable progress, current methods often present visual artifacts and can only deal with low-resolution images. In order to achieve aging/de-aging with the high quality and robustness necessary for wider use, these problems need to be addressed. This is the goal of the present work. We present an encoder-decoder architecture for face age editing. The core idea of our network is to encode a face image to age-invariant features, and learn a modulation vector corresponding to a target age. We then combine these two elements to produce a realistic image of the person with the desired target age. Our architecture is greatly simplified with respect to other approaches, and allows for fine-grained age editing on high resolution images in a single unified model. Source codes are available at https://github.com/InterDigitalInc/HRFAE.

We present a system for the removal of objects from videos. As input, the system only needs a user to draw a few strokes on the first frame, roughly delimiting the objects to be removed. To the best of our knowledge, this is the first system allowing the semi-automatic removal of objects from videos with complex backgrounds. The key steps of our system are the following: after initialization, segmentation masks are first refined and then automatically propagated through the video. Missing regions are then synthesized using video inpainting techniques. Our system can deal with multiple, possibly crossing objects, with complex motions, and with dynamic textures. This results in a computational tool that can alleviate tedious manual operations for editing high-quality videos.

Change detection is a key problem for many remote sensing applications. In this paper, we present a novel unsupervised method for change detection between two high-resolution remote sensing images possibly acquired by two different sensors. This method is based on keypoints matching, evaluation, and grouping, and does not require any image co-registration. It consists of two main steps. First, global and local mapping functions are estimated through keypoints extraction and matching. Second, based on these mappings, keypoint matchings are used to detect changes and then grouped to extract regions of changes. Both steps are defined through an a contrario framework, simplifying the parameter setting and providing a robust pipeline. The proposed approach is evaluated on synthetic and real data from different optic sensors with different resolutions, incidence angles, and illumination conditions.

In order to overcome the limited depth of field of usual photographic devices, a common approach is multi-focus image fusion (MFIF). From a stack of images acquired with different focus settings, these methods aim at fusing the content of the images of the stack to produce a final image that is sharp everywhere. Such methods can be very efficient, but when a global geometric alignment of images is out-of-reach, or when some objects are moving, the final image shows ghosts or other artefacts. In this paper, we propose a generic method to overcome these limitations. We first select a reference image, and then, for each image of the stack, reconstruct an image that shares the geometry of the reference and the sharpness content of the image at hand. The reconstruction is achieved thanks to a specially crafted modification of the PatchMatch algorithm, adapted to blurred images, and to a dedicated postprocessing for correcting reconstruction errors. Then, from the new image stack, MFIF is performed to produce a sharp result. We show the efficiency of the result on a database of challenging cases of hand-held shots containing moving objects.

This paper describes a simple and fast algorithm for removing occlusions that may occur in multiple views of a scene. In contrast to many methods of the literature, no assumption is made on occlusion shapes, colors or motions. Instead, this new method assumes that the background can be re-warped using an homography and that the reflectivity is quasi-Lambertian. After geometric and photometric alignments, three methods are evaluated. A median based method, a novel algorithm based on maximal clique detection and a robust PCA method are compared on real and simulated image sequences. This comparison shows that the new clique-based method provides the best performance in terms of quality and reliability.

The Copernicus Sentinel-2 program now provides multispectral images at a global scale with a high revisit rate. In this paper we explore the usage of convolutional neural networks for urban change detection using such multispectral images. We first present the new change detection dataset that was used for training the proposed networks, which will be openly available to serve as a benchmark. The Onera Satellite Change Detection (OSCD) dataset is composed of pairs of multispectral aerial images, and the changes were manually annotated at pixel level. We then propose two architectures to detect changes, Siamese and Early Fusion, and compare the impact of using different numbers of spectral channels as inputs. These architectures are trained from scratch using the provided dataset.

In this paper we propose to adapt LSD [1] (a state-of-the-art line segment detector for optical images) to SAR images. The first modification is replacing the gradient computation with an exponentially weighted ratio-based method which has a constant false alarm rate for SAR images. Next, we observe that the strong noise removal necessary for processing SAR images strongly impairs the independent hypothesis of the a contrario model used by LSD. A first order Markov chain is used to take the spatial dependencies into consideration. Experiments show that the proposed method has good performances and the number of false detections is well controlled.

We propose a fast and automatic inpainting technique for high-definition videos which works under many challenging conditions such as a moving camera, a dynamic background or a long lasting occlusion. Built upon the previous work by Newson et al. [1] which optimizes a global patch-based function, our method makes a significant improvement, especially in motion preservation, by incorporating the optical flow in several stages of the algorithm. Moreover, code parallelization and a modification in the process of patches pairwise matching yield a significant reduction of computation time. Experimental results on both classical and challenging datasets show that our algorithm outperforms other state-of-the-art approaches.

This demonstration aims to show some resulting videos for our method presented in [1]. It is a fast and automatic inpainting technique for high-definition videos which works under many challenging conditions such as a moving camera, a dynamic background or a long-lasting occlusion. By incorporating optical flow in a global patch-based algorithm, our method provide improvements compared to the state-of-the-art, especially in motion preservation. The aim of the demonstration is to provide the audience with inpainted high resolution videos, in which some objects are removed with hardly any remaining trace of their former presence. As such, it provides an interesting complement to the conference paper.

This paper presents a significant improvement for the synthesis of texture images using convolutional neural networks (CNNs), making use of constraints on the Fourier spectrum of the results. More precisely, the texture synthesis is regarded as a constrained optimization problem, with constraints conditioning both the Fourier spectrum and statistical features learned by CNNs. In contrast with existing methods, the presented method inherits from previous CNN approaches the ability to depict local structures and fine scale details, and at the same time yields coherent large scale structures, even in the case of quasi-periodic images. This is done at no extra computational cost. Synthesis experiments on various images show a clear improvement compared to a recent state-of-the art method relying on CNN constraints only.

This paper presents a novel variational approach to imposing statistical constraints on the output of both image generation (typically to perform texture synthesis) and image restoration (for instance, to achieve denoising and inpainting) methods. The empirical distributions of linear or nonlinear image descriptors are imposed to be close to some input distributions by minimizing a Wasserstein loss, i.e., the optimal transport distance between the distributions. We advocate the use of a Wasserstein distance because it is robust when using discrete distributions without the need to resort to kernel estimators. We showcase the use of different descriptors to tackle various image processing applications. These descriptors include linear wavelet-based filtering to account for simple textures, nonlinear sparse coding coefficients for more complicated patterns, and the image gradient to restore sharper contents. For applications to texture synthesis, the input distributions are the empirical distributions computed from an exemplar image. For image denoising and inpainting, the estimation process is more difficult; we propose making use of parametric models, and we show results using generalized Gaussian distributions.

The scale-invariant feature transform (SIFT) algorithm and its many variants are widely used in computer vision and in remote sensing to match features between images or to localize and recognize objects. However, mostly because of speckle noise, it does not perform well on synthetic aperture radar (SAR) images. In this paper, we introduce a SIFT-like algorithm specifically dedicated to SAR imaging, which is named SAR-SIFT. The algorithm includes both the detection of keypoints and the computation of local descriptors. A new gradient definition, yielding an orientation and a magnitude that are robust to speckle noise, is first introduced. It is then used to adapt several steps of the SIFT algorithm to SAR images. We study the improvement brought by this new algorithm, as compared with existing approaches. We present an application of SAR-SIFT to the registration of SAR images in different configurations, particularly with different incidence angles.

This paper introduces a new approach for the automatic estimation of illuminants in a digital color image. The method relies on two assumptions. First, the image is supposed to contain at least a small set of achromatic pixels. The second assumption is physical and concerns the set of possible illuminants, assumed to be well approximated by black body radiators. The proposed scheme is based on a projection of selected pixels on the Planckian locus in a well chosen chromaticity space, followed by a voting procedure yielding the estimation of the illuminant. This approach is very simple and learning-free. The voting procedure can be extended for the detection of multiple illuminants when necessary. Experiments on various databases show that the performances of this approach are similar to those of the best learning-based state-of-the-art algorithms.

Building high dynamic range (HDR) images by combining photographs captured with different exposure times present several drawbacks, such as the need for global alignment and motion estimation in order to avoid ghosting artifacts. The concept of spatially varying pixel exposures (SVE) proposed by Nayar et al. enables to capture in only one shot a very large range of exposures while avoiding these limitations. In this paper, we propose a novel approach to generate HDR images from a single shot acquired with spatially varying pixel exposures. The proposed method makes use of the assumption stating that the distribution of patches in an image is well represented by a Gaussian Mixture Model. Drawing on a precise modeling of the camera acquisition noise, we extend the piecewise linear estimation strategy developed by Yu et al. for image restoration. The proposed method permits to reconstruct an irradiance image by simultaneously estimating saturated and under-exposed pixels and denoising existing ones, showing significant improvements over existing approaches.

In disaster situations, remote sensing images are very useful to quickly assess damages. However, the choice of available images for the studied area is frequently limited. It is often needed to compare images acquired by different sensors and with different acquisition conditions. We propose a new feature-based approach to detect changes between a pair of either optical or radar images. This approach is based on the SIFT algorithm and an a contrario approach. It can deal with multi-resolutions, multi-sensors and multi-incidence angles situations, and it offers promising results.

Since the seminal work of Mann and Picard in 1995, the standard way to build high dynamic range (HDR) images from regular cameras has been to combine a reduced number of photographs captured with different exposure times. The algorithms proposed in the literature differ in the strategy used to combine these frames. Several experimental studies comparing their performances have been reported, showing in particular that a maximum likelihood estimation yields the best results in terms of mean squared error. However, no theoretical study aiming at establishing the performance limits of the HDR estimation problem has been conducted. Another common aspect of all HDR estimation approaches is that they discard saturated values. In this paper, we address these two issues. More precisely, we derive theoretical bounds for the performance of unbiased estimators for the HDR estimation problem. The unbiasedness hypothesis is motivated by the fact that most of the existing estimators, among them the best performing and most well known, are nearly unbiased. Moreover, we show that, even with a small number of photographs, the maximum likelihood estimator performs extremely close to these bounds. As a second contribution, we propose a general strategy for integrating the information provided by saturated pixels in the estimation process, hence improving the estimation results. Finally, we analyze the sensitivity of the HDR estimation process to camera parameters, and we show that small errors in the camera calibration process may severely degrade the estimation results.

We propose an automatic video inpainting algorithm which relies on the optimization of a global, patch-based functional. Our algorithm is able to deal with a variety of challenging situations which naturally arise in video inpainting, such as the correct reconstruction of dynamic textures, multiple moving objects, and moving background. Furthermore, we achieve this in an order of magnitude less execution time with respect to the state-of-the-art. We are also able to achieve good quality results on high-definition videos. Finally, we provide specific algorithmic details to make implementation of our algorithm as easy as possible. The resulting algorithm requires no segmentation or manual input other than the definition of the inpainting mask and can deal with a wider variety of situations than is handled by previous work.

Line scratch detection in old films is a particularly challenging problem due to the variable spatiotemporal characteristics of this defect. Some of the main problems include sensitivity to noise and texture, and false detections due to thin vertical structures belonging to the scene. We propose a robust and automatic algorithm for frame-by-frame line scratch detection in old films, as well as a temporal algorithm for the filtering of false detections. In the frame-by-frame algorithm, we relax some of the hypotheses used in previous algorithms in order to detect a wider variety of scratches. This step's robustness and lack of external parameters is ensured by the combined use of an a contrario methodology and local statistical estimation. In this manner, over-detection in textured or cluttered areas is greatly reduced. The temporal filtering algorithm eliminates false detections due to thin vertical structures by exploiting the coherence of their motion with that of the underlying scene. Experiments demonstrate the ability of the resulting detection procedure to deal with difficult situations, in particular in the presence of noise, texture, and slanted or partial scratches. Comparisons show significant advantages over previous work.

High dynamic range (HDR) images are usually generated by combining multiple photographs acquired with different exposure times. This approach, while effective, suffers from various drawbacks. The irradiance estimation is performed by combining, for each pixel, different exposure values at the same spatial position. This estimation scheme does not take advantage of the redundancy present in most images. Moreover, images must be perfectly aligned and objects must be in the exact same position in all frames in order to combine the different exposures. In this work, we propose a new HDR image generation approach that simultaneously copes with these problems and exploits image redundancy to produce a denoised result. A reference image is chosen and a patch-based approach is used to find similar pixels that are then combined for the irradiance estimation. This patch-based approach permits to obtain a denoised result and is robust to image misalignments and object motions. Results show significant improvements in terms of noise reduction over previous HDR image generation techniques, while being robust to motion and changes between the exposures.

The film defect known as the line scratch is difficult to restore automatically due to the large number of false alarms present in scratch detection algorithms. In this paper, an algorithm for dealing with these false alarms is proposed. Validating true scratches, which is the approach generally proposed in the literature, is a difficult task since scratch characteristics are hard to determine, making tracking these defects problematic. Instead, we eliminate false alarms by analysing their compatibility with a global motion estimation. We compare our algorithm with two other scratch detection methods from the literature. Experiments show that our algorithm outperforms these two, and that the proposed temporal filtering greatly improves precision while maintaining high recall.

The scale invariant feature transform (SIFT) algorithm, commonly used in computer vision, does not perform well on synthetic aperture radar (SAR) images, in particular because of the strong intensity and the multiplicative nature of the noise. We present an improvement of this algorithm for SAR images. First, a robust yet simple way to compute gradient on radar images is introduced. This step is first used to develop a new keypoints extraction algorithm, based on the Harris criterion. Second, we rely on this gradient definition to adapt the computation of both the main orientation and the geometric descriptor to SAR image specificities. We validate this new algorithm with different experiments and present an application of our new SAR-SIFT algorithm.

Exemplar-based methods have proven their efficiency for the reconstruction of missing parts in a digital image. Texture as well as local geometry are often very well restored by such methods. Some applications, however, require the ability to reconstruct nonlocal geometric features, e.g., long edges. In order to do so, we propose to first compute a geometric sketch, which is then interpolated and used as a guide for the global reconstruction. In comparison with other related approaches, the originality of our work relies on the following points: (1) The geometric sketch computation is parameter-free and based on level lines, which provides a complete, reliable, and stable representation of the image. (2) The completion of the geometric sketch is fully automatic. It is done using a new—and interesting on its own—geometric inpainting approach that interpolates level lines with Euler spirals. Euler spirals are natural curves for shape completion and have been used already for edge completion and inpainting. It is the first time, however, that these curves are used for completing the whole level lines structure. (3) The general reconstruction is performed using a guided version of a classical exemplar-based method. However, we do not constrain the exemplar-based reconstruction to strictly follow the geometric guide. We actually use a new metric between blocks that consists of the sum of the classical ${{\mathrm L}^2}$ metric between any two blocks of the general image plus an ${{\mathrm L}^2}$ metric between the corresponding blocks in the completed geometric image. This is equivalent to a Lagrangian relaxation of a strictly guided reconstruction. We discuss in the paper the details of the method and some related mathematical issues, and we illustrate its efficiency on several examples.

This paper deals with the parameter choice for the nonlocal means (NLM) algorithm. After basic computations on toy models highlighting the bias of the NLM, we study the bias-variance trade-off of this filter so as to highlight the need of a local choice of the parameters. Relying on Stein's unbiased risk estimate, we then propose an efficient algorithm to locally set these parameters, and we compare this method with the NLM with optimal global parameter.

This paper explores the mathematical and algorithmic properties of two sample-based texture models: random phase noise (RPN) and asymptotic discrete spot noise (ADSN). These models permit to synthesize random phase textures. They arguably derive from linearized versions of two early Julesz texture discrimination theories. The ensuing mathematical analysis shows that, contrarily to some statements in the literature, RPN and ADSN are different stochastic processes. Nevertheless, numerous experiments also suggest that the textures obtained by these algorithms from identical samples are perceptually similar. The relevance of this study is enhanced by three technical contributions providing solutions to obstacles that prevented the use of RPN or ADSN to emulate textures. First, RPN and ADSN algorithms are extended to color images. Second, a preprocessing is proposed to avoid artifacts due to the nonperiodicity of real-world texture samples. Finally, the method is extended to synthesize textures with arbitrary size from a given sample.

This work is concerned with the modification of the gray level or color distribution of digital images. A common drawback of classical methods aiming at such modifications is the revealing of artefacts or the attenuation of details and textures. In this work, we propose a generic filtering method enabling, given the original image and the radiometrically corrected one, to suppress artefacts while preserving details. The approach relies on the key observation that artefacts correspond to spatial irregularity of the so-called transportation map, defined as the difference between the original and the corrected image. The proposed method draws on the nonlocal Yaroslavsky filter to regularize the transportation map. The efficiency of the method is shown on various radiometric modifications: contrast equalization, midway histogram, color enhancement, and color transfer. A comparison with related approaches is also provided.

Blotches are very common, localized, and non persistent impairments in digitized film archive. Many methods have been proposed so far for detecting them and restoring the underlying regions. Most detection techniques rely on the hypothesis that blotches contradict a model of motion regularity and, up to a prior motion compensation, correspond to significant temporal variations of intensity with respect to a global threshold. In this paper, we propose a statistical approach to detect blotches in image sequences, which yields thresholds adapted to the local statistics of the frames, and which takes into account gray level differences in neighborhoods instead of isolated points. This approach is combined with a block-based motion estimation. The whole procedure is confronted with classical approaches on several sequences.

In this paper, we take interest in the process of assigning a given color distribution to an image. Two examples of such image modifications are histogram equalization (or specification) and color transfer, in which the color palette of a style image is assigned to a source image. Classical methods for gray level specification, as well as more recent methods for color transfer, can be defined as optimal transportation problems. The corresponding image modifications are known to produce visually unpleasing effects such as the removal of details and texture, as well as the enhancement of noise or compression patterns. In this paper, a new method is proposed for the suppression of these artifacts. The method relies on a non local regularization of the transportation map, defined as the difference between the original image and the modified one. The interest of using this method is demonstrated on the aforementioned applications: contrast adjustment and color transfer.

This paper focuses on the matching of local features between images. Given a set of query descriptors and a database of candidate descriptors, the goal is to decide which ones should be matched. This is a crucial issue, since the matching procedure is often a preliminary step for object detection or image matching. In practice, this matching step is often reduced to a specific threshold on the Euclidean distance to the nearest neighbor. Our first contribution is a robust distance between descriptors, relying on the adaptation of the Earth Mover's Distance to circular histograms. It is shown that this distance outperforms classical distances for comparing SIFT-like descriptors, while its time complexity remains reasonable. Our second and main contribution is a statistical framework for the matching procedure, which yields validation thresholds automatically adapted to the complexity of each query descriptor and to the diversity and size of the database. The method makes it possible to detect multiple occurrences, as well as to deal with situations where the target is not present. Its performances are tested through various experiments on a large image database.

In this paper, the matching of SIFT-like features [5] between images is studied. The goal is to decide which matches between descriptors of two datasets should be selected. This matching procedure is often a preliminary step towards some computer vision applications, such as object detection and image registration for instance. The distances between the query descriptors and the database candidates being computed, the classical approach is to select for each query its nearest neighbor, depending on a global threshold on dissimilarity measure. In this contribution, an a contrario framework for the matching procedure is introduced, based on a threshold on a probability of false detections. This approach yields dissimilarity thresholds automatically adapted to each query descriptor and to the diversity and size of the database. We show on various experiments on a large image database, the ability of such a method to decide whether a query and its candidates should be matched.

Many computer vision algorithms make use of local features, and rely on a systematic comparison of these features. The chosen dissimilarity measure is of crucial importance for the overall performances of these algorithms and has to be both robust and computationally efficient. Some of the most popular local features (like SIFT [4] descriptors) are based on one-dimensional circular histograms. In this contribution, we present an adaptation of the Earth moverpsilas distance to one-dimensional circular histograms. This distance, that we call CEMD, is used to compare SIFT-like descriptors. Experiments over a large database of 3 million descriptors show that CEMD outperforms classical bin-to-bin distances, while having reasonable time complexity.

In this paper, we present a set of texture features that are locally invariant to similarity or affinity. The proposed indexing scheme relies on the topographic map, a shape-based representation of images. Thanks to the hierarchical organization of the topographic map, the approach gives a grip on the multi-scale structure of textures. Using simple one dimensional histograms, the method is shown to achieve state-of-the-art performances among locally invariant methods, both on the whole Brodatz and UIUC databases.

Space agencies are rapidly building up massive image databases. A particularity of these databases is that they are made of images with different, but known, resolutions. In this paper, we introduce a new scheme allowing us to compare and index images with different resolutions. This scheme relies on a simplified acquisition model of satellite images and uses continuous wavelet decompositions. We establish a correspondence between scales which permits us to compare wavelet decompositions of images having different resolutions. We validate the approach through several matching and classification experiments, and we show that taking the acquisition process into account yields better results than just using scaling properties of wavelet features.

We study the problem of finding the characteristic scale of a given satellite image. This feature is defined so that it does not depend on the spatial resolution of the image. This is a different problem than achieving scale invariance, as often studied in the literature. Our approach is based on the use of a linear scale space and the total variation (TV). The critical scale is defined as the one at which the normalized TV reaches its maximum. It is shown experimentally, both on synthetic and real data, that the computed characteristic scale is resolution independent.

We study the problem of finding the characteristic scale of a given satellite image. We want to define this feature so that it does not depend on the spatial resolution of the image. Our approach is based on the use of a linear scale space and the total variation. The critical scale is defined as the one at which the normalized total variation is maximum.

In this paper, we propose a new scheme to extrapolate wavelet features with respect to the resolution. By explicitly taking into account the acquisition process of satellite images, we compute how wavelet features behave when the resolution changes. This approach is validated by classifying satellite images with different resolutions.

Shape recognition systems usually order a fixed number of best matches to each query, but do not address or answer the two following questions: Is a query shape in a given database? How can we be sure that a match is correct? This communication deals with these two key points. A database being given, with each shape S and each distance /spl delta/, we associate its number of false alarms NFA(S, /spl delta/), namely the expectation of the number of shapes at distance /spl delta/ in the database. Assume that NFA(S, /spl delta/) is very small with respect to 1, and that a shape S' is found at distance /spl delta/ from S in the database. This match could not occur just by chance and is therefore a meaningful detection. Its explanation is usually the common origin of both shapes. Experimental evidence will show that NFA(S, /spl delta/) can be predicted accurately.

Interpolation of digital elevation models becomes necessary in many situations, for instance, when constructing them from contour lines (available e.g., from nondigital cartography), or from disparity maps based on pairs of stereoscopic views, which often leaves large areas where point correspondences cannot be found reliably. The absolutely minimizing Lipschitz extension (AMLE) model is singled out as the simplest interpolation method satisfying a set of natural requirements. In particular, a maximum principle is proven, which guarantees not to introduce unnatural oscillations which is a major problem with many classical methods. The authors then discuss the links between the AMLE and other existing methods. In particular, they show its relation with geodesic distance transformation. They also relate the AMLE to the thin-plate method, that can be obtained by a prolongation of the axiomatic arguments leading to the AMLE, and addresses the major disadvantage of the AMLE model, namely its inability to interpolate slopes as it does for values. Nevertheless, in order to interpolate slopes, they have to give up the maximum principle and authorize the appearance of oscillations. They also discuss the possible link between the AMLE and Kriging methods that are the most widely used in the geoscience literature.