Landuse characterization is important for urban planning. It is traditionally performed with field surveys or manual photo interpretation, two practices that are time-consuming and labor-intensive. Therefore, we aim to automate landuse mapping at the urban-object level with a deep learning approach based on data from multiple sources (or modalities). We consider two image modalities: overhead imagery from Google Maps and ensembles of ground-based pictures (side-views) per urban-object from Google Street View (GSV). These modalities bring complementary visual information pertaining to the urban-objects. We propose an end-to-end trainable model, which uses OpenStreetMap annotations as labels. The model can accommodate a variable number of GSV pictures for the ground-based branch and can also function in the absence of ground pictures at prediction time. We test the effectiveness of our model over the area of \^Ile-de-France, France, and test its generalization abilities on a set of urban-objects from the city of Nantes, France. Our proposed multimodal Convolutional Neural Network achieves considerably higher accuracies than methods that use a single image modality, making it suitable for automatic landuse map updates. Additionally, our approach could be easily scaled to multiple cities, because it is based on data sources available for many cities worldwide.
Noisy labels often occur in vision datasets, especially when they are issued from crowdsourcing or Web scraping. In this paper, we propose a new regularization method which enables one to learn robust classifiers in presence of noisy data. To achieve this goal, we augment the virtual adversarial loss with a Wasserstein distance. This distance allows us to take into account specific relations between classes by leveraging on the geometric properties of this optimal transport distance. Notably, we encode the class similarities in the ground cost that is used to compute the Wasserstein distance. As a consequence, we can promote smoothness between classes that are very dissimilar, while keeping the classification decision function sufficiently complex for similar classes. While designing this ground cost can be left as a problem-specific modeling task, we show in this paper that using the semantic relations between classes names already leads to good results.Our proposed Wasserstein Adversarial Training (WAT) outperforms state of the art on four datasets corrupted with noisy labels: three classical benchmarks and one real case in remote sensing image semantic segmentation.
Rural building mapping is paramount to support demographic studies and plan actions in response to crisis that affect those areas. Rural building annotations exist in OpenStreetMap (OSM), but their quality and quantity are not sufficient for training models that can create accurate rural building maps. The problems with these annotations essentially fall into three categories: (i) most commonly, many annotations are geometrically misaligned with the updated imagery; (ii) some annotations do not correspond to buildings in the images (they are misannotations or the buildings have been destroyed); and (iii) some annotations are missing for buildings in the images (the buildings were never annotated or were built between subsequent image acquisitions). First, we propose a method based on Markov Random Field (MRF) to align the buildings with their annotations. The method maximizes the correlation between annotations and a building probability map while enforcing that nearby buildings have similar alignment vectors. Second, the annotations with no evidence in the building probability map are removed. Third, we present a method to detect non-annotated buildings with predefined shapes and add their annotation. The proposed methodology shows considerable improvement in accuracy of the OSM annotations for two regions of Tanzania and Zimbabwe, being more accurate than state-of-the-art baselines.
In computer vision, one is often confronted with problems of domain shifts, which occur when one applies a classifier trained on a source dataset to target data sharing similar characteristics (e.g. same classes), but also different latent data structures (e.g. different acquisition conditions). In such a situation, the model will perform poorly on the new data, since the classifier is specialized to recognize visual cues specific to the source domain. In this work we explore a solution, named DeepJDOT, to tackle this problem: through a measure of discrepancy on joint deep representations/labels based on optimal transport, we not only learn new data representations aligned between the source and target domain, but also simultaneously preserve the discriminative information used by the classifier. We applied DeepJDOT to a series of visual recognition tasks, where it compares favorably against state-of-the-art deep domain adaptation methods.
Classification of remotely sensed images into land cover or land use is highly dependent on geographical information at least at two levels. First, land cover classes are observed in a spatially smooth domain separated by sharp region boundaries. Second, land classes and observation scale are also tightly intertwined: they tend to be consistent within areas of homogeneous appearance, or regions, in the sense that all pixels within a roof should be classified as roof, independently on the spatial support used for the classification. In this paper, we follow these two observations and encode them as priors in an energy minimization framework based on conditional random fields (CRFs), where classification results obtained at pixel and region levels are probabilistically fused. The aim is to enforce the final maps to be consistent not only in their own spatial supports (pixel and region) but also across supports, i.e., by getting the predictions on the pixel lattice and on the set of regions to agree. To this end, we define an energy function with three terms: 1) a data term for the individual elements in each support (support-specific nodes); 2) spatial regularization terms in a neighborhood for each of the supports (support-specific edges); and 3) a regularization term between individual pixels and the region containing each of them (intersupports edges). We utilize these priors in a unified energy minimization problem that can be optimized by standard solvers. The proposed 2LCRF model consists of a CRF defined over a bipartite graph, i.e., two interconnected layers within a single graph accounting for interlattice connections.
When approaching the semantic segmentation of overhead imagery in the decimeter spatial resolution range, successful strategies usually combine powerful methods to learn the visual appearance of the semantic classes (e.g. convolutional neural networks) with strategies for spatial regularization (e.g. graphical models such as conditional random fields). In this paper, we propose a method to learn evidence in the form of semantic class likelihoods, semantic boundaries across classes and shallow-to-deep visual features, each one modeled by a multi-task convolutional neural network architecture. We combine this bottom-up information with top-down spatial regularization encoded by a conditional random field model optimizing the label space across a hierarchy of segments with constraints related to structural, spatial and data-dependent pairwise relationships between regions. Our results show that such strategy provide better regularization than a series of strong baselines reflecting state-of-the-art technologies. The proposed strategy offers a flexible and principled framework to include several sources of visual and structural information, while allowing for different degrees of spatial regularization accounting for priors about the expected output structures.
We study the effect of injecting local scale equivariance into Convolutional Neural Networks. This is done by applying each convolutional filter at multiple scales. The output is a vector field encoding for the maximally activating scale and the scale itself, which is further processed by the following convolutional layers. This allows all the intermediate representations to be locally scale equivariant. We show that this improves the performance of the model by over $20\%$ in the scale equivariant task of regressing the scaling factor applied to randomly scaled MNIST digits. Furthermore, we find it also useful for scale invariant tasks, such as the actual classification of randomly scaled digits. This highlights the usefulness of allowing for a compact representation that can also learn relationships between different local scales by keeping internal scale equivariance.
Detecting topographic changes in the urban environment has always been an important task for urban planning and monitoring. In practice, remote sensing data are often available in different modalities and at different time epochs. Change detection between multimodal data can be very challenging since the data show different characteristics. Given 3D laser scanning point clouds and 2D imagery from different epochs, this paper presents a framework to detect building and tree changes. First, the 2D and 3D data are transformed to image patches, respectively. A Siamese CNN is then employed to detect candidate changes between the two epochs. Finally, the candidate patch-based changes are grouped and verified as individual object changes. Experiments on the urban data show that 86.4\% of patch pairs can be correctly classified by the model.
Knowledge over the number of animals in large wildlife reserves is a vital necessity for park rangers in their efforts to protect endangered species. Manual animal censuses are dangerous and expensive, hence Unmanned Aerial Vehicles (UAVs) with consumer level digital cameras are becoming a popular alternative tool to estimate livestock. Several works have been proposed that semi-automatically process UAV images to detect animals, of which some employ Convolutional Neural Networks (CNNs), a recent family of deep learning algorithms that proved very effective in object detection in large datasets from computer vision. However, the majority of works related to wildlife focuses only on small datasets (typically subsets of UAV campaigns), which might be detrimental when presented with the sheer scale of real study areas for large mammal census. Methods may yield thousands of false alarms in such cases. In this paper, we study how to scale CNNs to large wildlife census tasks and present a number of recommendations to train a CNN on a large UAV dataset. We further introduce novel evaluation protocols that are tailored to censuses and model suitability for subsequent human verification of detections. Using our recommendations, we are able to train a CNN reducing the number of false positives by an order of magnitude compared to previous state-of-the-art. Setting the requirements at 90% recall, our CNN allows to reduce the amount of data required for manual verification by three times, thus making it possible for rangers to screen all the data acquired efficiently and to detect almost all animals in the reserve automatically.
In this paper, we propose to tackle the problem of reducing discrepancies between multiple domains referred to as multi-source domain adaptation and consider it under the target shift assumption: in all domains we aim to solve a classification problem with the same output classes, but with labels' proportions differing across them. This problem, generally ignored in the vast majority papers on domain adaptation papers, is nevertheless critical in real-world applications, and we theoretically show its impact on the adaptation success. To address this issue, we design a method based on optimal transport, a theory that has been successfully used to tackle adaptation problems in machine learning. Our method performs multi-source adaptation and target shift correction simultaneously by learning the class probabilities of the unlabeled target sample and the coupling allowing to align two (or more) probability distributions. Experiments on both synthetic and real-world data related to satellite image segmentation task show the superiority of the proposed method over the state-of-the-art.