Abstract: Human visual perception is limited to the visual-optical spectrum. Machine vision is not. Cameras sensitive to the different infrared spectra can enhance the abilities of autonomous systems and visually perceive the environment in a holistic way. Relevant scene content can be made visible especially in situations, where sensors of other modalities face issues like a visual-optical camera that needs a source of illumination. As a consequence, not only human mistakes can be avoided by increasing the level of automation, but also machine-induced errors can be reduced that, for example, could make a self-driving car crash into a pedestrian under difficult illumination conditions. Furthermore, multi-spectral sensor systems with infrared imagery as one modality are a rich source of information and can provably increase the robustness of many autonomous systems. Applications that can benefit from utilizing infrared imagery range from robotics to automotive and from biometrics to surveillance. In this book, we provide a brief yet concise introduction to the current state-of-the-art of computer vision and machine learning in the infrared spectrum. Based on various popular computer vision tasks such as image enhancement, object detection, or object tracking, we first motivate each task starting from established literature in the visual-optical spectrum. Then, we discuss the differences between processing images and videos in the visual-optical spectrum and the various infrared spectra. An overview of the current literature is provided together with an outlook for each task. Furthermore, available and annotated public datasets and common evaluation methods and metrics are presented. In a separate chapter, popular applications that can greatly benefit from the use of infrared imagery as a data source are presented and discussed. Among them are automatic target recognition, video surveillance, or biometrics including face recognition. Finally, we conclude with recommendations for well-fitting sensor setups and data processing algorithms for certain computer vision tasks. We address this book to prospective researchers and engineers new to the field but also to anyone who wants to get introduced to the challenges and the approaches of computer vision using infrared images or videos. Readers will be able to start their work directly after reading the book supported by a highly comprehensive backlog of recent and relevant literature as well as related infrared datasets including existing evaluation frameworks. Together with consistently decreasing costs for infrared cameras, new fields of application appear and make computer vision in the infrared spectrum a great opportunity to face nowadays scientific and engineering challenges.

Abstract: Lateral connections in the primary visual cortex (V1) have long been hypothesized to be responsible for several visual processing mechanisms such as brightness induction, chromatic induction, visual discomfort, and bottom-up visual attention (also named saliency). Many computational models have been developed to independently predict these and other visual processes, but no computational model has been able to reproduce all of them simultaneously. In this work, we show that a biologically plausible computational model of lateral interactions of V1 is able to simultaneously predict saliency and all the aforementioned visual processes. Our model's architecture (NSWAM) is based on Penacchio's neurodynamic model of lateral connections of V1. It is defined as a network of firing rate neurons, sensitive to visual features such as brightness, color, orientation, and scale. We tested NSWAM saliency predictions using images from several eye tracking data sets. We show that the accuracy of predictions obtained by our architecture, using shuffled metrics, is similar to other state-of-the-art computational methods, particularly with synthetic images (CAT2000-Pattern and SID4VAM) that mainly contain low-level features. Moreover, we outperform other biologically inspired saliency models that are specifically designed to exclusively reproduce saliency. We show that our biologically plausible model of lateral connections can simultaneously explain different visual processes present in V1 (without applying any type of training or optimization and keeping the same parameterization for all the visual processes). This can be useful for the definition of a unified architecture of the primary visual cortex.

Abstract: Semantic image segmentation is a central and challenging task in autonomous driving, addressed by training deep models. Since this training draws to a curse of human-based image labeling, using synthetic images with automatically generated labels together with unlabeled real-world images is a promising alternative. This implies to address an unsupervised domain adaptation (UDA) problem. In this paper, we propose a new co-training procedure for synth-to-real UDA of semantic
segmentation models. It consists of a self-training stage, which provides two domain-adapted models, and a model collaboration loop for the mutual improvement of these two models. These models are then used to provide the final semantic segmentation labels (pseudo-labels) for the real-world images. The overall
procedure treats the deep models as black boxes and drives their collaboration at the level of pseudo-labeled target images, i.e., neither modifying loss functions is required, nor explicit feature alignment. We test our proposal on standard synthetic and real-world datasets for on-board semantic segmentation. Our
procedure shows improvements ranging from ∼13 to ∼26 mIoU points over baselines, so establishing new state-of-the-art results.

Abstract: In this paper, we propose a Text-Degradation Invariant Auto Encoder (Text-DIAE), a self-supervised model designed to tackle two tasks, text recognition (handwritten or scene-text) and document image enhancement. We start by employing a transformer-based architecture that incorporates three pretext tasks as learning objectives to be optimized during pre-training without the usage of labelled data. Each of the pretext objectives is specifically tailored for the final downstream tasks. We conduct several ablation experiments that confirm the design choice of the selected pretext tasks. Importantly, the proposed model does not exhibit limitations of previous state-of-the-art methods based on contrastive losses, while at the same time requiring substantially fewer data samples to converge. Finally, we demonstrate that our method surpasses the state-of-the-art in existing supervised and self-supervised settings in handwritten and scene text recognition and document image enhancement. Our code and trained models will be made publicly available at https://github.com/dali92002/SSL-OCR

Abstract: Global visual localization estimates the absolute pose of a camera using a single image, in a previously mapped area. Obtaining the pose from a single image enables many robotics and augmented/virtual reality applications. Inspired by latest advances in deep learning, many existing approaches directly learn and regress 6 DoF pose from an input image. However, these methods do not fully utilize the underlying scene geometry for pose regression. The challenge in monocular relocalization is the minimal availability of supervised training data, which is just the corresponding 6 DoF poses of the images. In this letter, we propose to utilize these minimal available labels (i.e., poses) to learn the underlying 3D geometry of the scene and use the geometry to estimate the 6 DoF camera pose. We present a learning method that uses these pose labels and rigid alignment to learn two 3D geometric representations ( X, Y, Z coordinates ) of the scene, one in camera coordinate frame and the other in global coordinate frame. Given a single image, it estimates these two 3D scene representations, which are then aligned to estimate a pose that matches the pose label. This formulation allows for the active inclusion of additional learning constraints to minimize 3D alignment errors between the two 3D scene representations, and 2D re-projection errors between the 3D global scene representation and 2D image pixels, resulting in improved localization accuracy. During inference, our model estimates the 3D scene geometry in camera and global frames and aligns them rigidly to obtain pose in real-time. We evaluate our work on three common visual localization datasets, conduct ablation studies, and show that our method exceeds state-of-the-art regression methods' pose accuracy on all datasets.

Abstract: Despite its wide applications in criminal investigations and clinical communications with patients suffering from autism, automatic micro-expression recognition remains a challenging problem because of the lack of training data and imbalanced classes problems. In this study, we proposed a meta-learning-based multi-model fusion network (Meta-MMFNet) to solve the existing problems. The proposed method is based on the metric-based meta-learning pipeline, which is specifically designed for few-shot learning and is suitable for model-level fusion. The frame difference and optical flow features were fused, deep features were extracted from the fused feature, and finally in the meta-learning-based framework, weighted sum model fusion method was applied for micro-expression classification. Meta-MMFNet achieved better results than state-of-the-art methods on four datasets. The code is available at https://github.com/wenjgong/meta-fusion-based-method.

Abstract: This paper deals with the recognition of symbols and structural textures in architectural plans using string matching techniques. A plan is represented by an attributed graph whose nodes represent characteristic points and whose edges represent segments. Symbols and textures can be seen as a set of regions, i.e. closed loops in the graph, with a particular arrangement. The search for a symbol involves a graph matching between the regions of a model graph and the regions of the graph representing the document. Discriminating a texture means a clustering of neighbouring regions of this graph. Both procedures involve a similarity measure between graph regions. A string codification is used to represent the sequence of outlining edges of a region. Thus, the similarity between two regions is defined in terms of the string edit distance between their boundary strings. The use of string matching allows the recognition method to work also under presence of distortion.