High-Resolution Hyperspectral Video Imaging Using A Hexagonal Camera Array

Retrieving the reflectance spectrum from objects is an essential task for many classification and detection problems, since many materials and processes have a unique spectral behaviour. In many cases, it is highly desirable to capture hyperspectral images due to the high spectral flexibility. Often, it is even necessary to capture hyperspectral videos or at least to be able to record a hyperspectral image at once, also called snapshot hyperspectral imaging, to avoid spectral smearing. For this task, a high-resolution snapshot hyperspectral camera array using a hexagonal shape is introduced.The hexagonal array for hyperspectral imaging uses off-the-shelf hardware, which enables high flexibility regarding employed cameras, lenses and filters. Hence, the spectral range can be easily varied by mounting a different set of filters. Moreover, the concept of using off-the-shelf hardware enables low prices in comparison to other approaches with highly specialized hardware. Since classical industrial cameras are used in this hyperspectral camera array, the spatial and temporal resolution is very high, while recording 37 hyperspectral channels in the range from 400 nm to 760 nm in 10 nm steps. A registration process is required for near-field imaging, which maps the peripheral camera views to the center view. It is shown that this combination using a hyperspectral camera array and the corresponding image registration pipeline is superior in comparison to other popular snapshot approaches. For this evaluation, a synthetic hyperspectral database is rendered. On the synthetic data, the novel approach outperforms its best competitor by more than 3 dB in reconstruction quality. This synthetic data is also used to show the superiority of the hexagonal shape in comparison to an orthogonal-spaced one. Moreover, a real-world high resolution hyperspectral video database is provided.

A Study on the Effect of Color Spaces in Learned Image Compression

In this work, we present a comparison between color spaces namely YUV, LAB, RGB and their effect on learned image compression. For this we use the structure and color based learned image codec (SLIC) from our prior work, which consists of two branches - one for the luminance component (Y or L) and another for chrominance components (UV or AB). However, for the RGB variant we input all 3 channels in a single branch, similar to most learned image codecs operating in RGB. The models are trained for multiple bitrate configurations in each color space. We report the findings from our experiments by evaluating them on various datasets and compare the results to state-of-the-art image codecs. The YUV model performs better than the LAB variant in terms of MS-SSIM with a Bj{\o}ntegaard delta bitrate (BD-BR) gain of 7.5\% using VTM intra-coding mode as the baseline. Whereas the LAB variant has a better performance than YUV model in terms of CIEDE2000 having a BD-BR gain of 8\%. Overall, the RGB variant of SLIC achieves the best performance with a BD-BR gain of 13.14\% in terms of MS-SSIM and a gain of 17.96\% in CIEDE2000 at the cost of a higher model complexity.

Multispectral Snapshot Image Registration Using Learned Cross Spectral Disparity Estimation and a Deep Guided Occlusion Reconstruction Network

Multispectral imaging aims at recording images in different spectral bands. This is extremely beneficial in diverse discrimination applications, for example in agriculture, recycling or healthcare. One approach for snapshot multispectral imaging, which is capable of recording multispectral videos, is by using camera arrays, where each camera records a different spectral band. Since the cameras are at different spatial positions, a registration procedure is necessary to map every camera to the same view. In this paper, we present a multispectral snapshot image registration with three novel components. First, a cross spectral disparity estimation network is introduced, which is trained on a popular stereo database using pseudo spectral data augmentation. Subsequently, this disparity estimation is used to accurately detect occlusions by warping the disparity map in a layer-wise manner. Finally, these detected occlusions are reconstructed by a learned deep guided neural network, which leverages the structure from other spectral components. It is shown that each element of this registration process as well as the final result is superior to the current state of the art. In terms of PSNR, our registration achieves an improvement of over 3 dB. At the same time, the runtime is decreased by a factor of over 3 on a CPU. Additionally, the registration is executable on a GPU, where the runtime can be decreased by a factor of 111. The source code and the data is available at https://github.com/FAU-LMS/MSIR.

A Guided Upsampling Network for Short Wave Infrared Images Using Graph Regularization

Exploiting the infrared area of the spectrum for classification problems is getting increasingly popular, because many materials have characteristic absorption bands in this area. However, sensors in the short wave infrared (SWIR) area and even higher wavelengths have a very low spatial resolution in comparison to classical cameras that operate in the visible wavelength area. Thus, in this paper an upsampling method for SWIR images guided by a visible image is presented. For that, the proposed guided upsampling network (GUNet) uses a graph-regularized optimization problem based on learned affinities is presented. The evaluation is based on a novel synthetic near-field visible-SWIR stereo database. Different guided upsampling methods are evaluated, which shows an improvement of nearly 1 dB on this database for the proposed upsampling method in comparison to the second best guided upsampling network. Furthermore, a visual example of an upsampled SWIR image of a real-world scene is depicted for showing real-world applicability.

Color Agnostic Cross-Spectral Disparity Estimation

Since camera modules become more and more affordable, multispectral camera arrays have found their way from special applications to the mass market, e.g., in automotive systems, smartphones, or drones. Due to multiple modalities, the registration of different viewpoints and the required cross-spectral disparity estimation is up to the present extremely challenging. To overcome this problem, we introduce a novel spectral image synthesis in combination with a color agnostic transform. Thus, any recently published stereo matching network can be turned to a cross-spectral disparity estimator. Our novel algorithm requires only RGB stereo data to train a cross-spectral disparity estimator and a generalization from artificial training data to camera-captured images is obtained. The theoretical examination of the novel color agnostic method is completed by an extensive evaluation compared to state of the art including self-recorded multispectral data and a reference implementation. The novel color agnostic disparity estimation improves cross-spectral as well as conventional color stereo matching by reducing the average end-point error by 41% for cross-spectral and by 22% for mono-modal content, respectively.

RGB-Guided Resolution Enhancement of IR Images

This paper introduces a novel method for RGB-Guided Resolution Enhancement of infrared (IR) images called Guided IR Resolution Enhancement (GIRRE). In the area of single image super resolution (SISR) there exists a wide variety of algorithms like interpolation methods or neural networks to improve the spatial resolution of images. In contrast to SISR, even more information can be gathered on the recorded scene when using multiple cameras. In our setup, we are dealing with multi image super resolution, especially with stereo super resolution. We consider a color camera and an IR camera. Current IR sensors have a very low resolution compared to color sensors so that recent color sensors take up 100 times more pixels than IR sensors. To this end, GIRRE increases the spatial resolution of the low-resolution IR image. After that, the upscaled image is filtered with the aid of the high-resolution color image. We show that our method achieves an average PSNR gain of 1.2 dB and at best up to 1.8 dB compared to state-of-the-art methods, which is visually noticeable.

Conditional Residual Coding: A Remedy for Bottleneck Problems in Conditional Inter Frame Coding

Conditional coding is a new video coding paradigm enabled by neural-network-based compression. It can be shown that conditional coding is in theory better than the traditional residual coding, which is widely used in video compression standards like HEVC or VVC. However, on closer inspection, it becomes clear that conditional coders can suffer from information bottlenecks in the prediction path, i.e., that due to the data processing inequality not all information from the prediction signal can be passed to the reconstructed signal, thereby impairing the coder performance. In this paper we propose the conditional residual coding concept, which we derive from information theoretical properties of the conditional coder. This coder significantly reduces the influence of bottlenecks, while maintaining the theoretical performance of the conditional coder. We provide a theoretical analysis of the coding paradigm and demonstrate the performance of the conditional residual coder in a practical example. We show that conditional residual coders alleviate the disadvantages of conditional coders while being able to maintain their advantages over residual coders. In the spectrum of residual and conditional coding, we can therefore consider them as ``the best from both worlds''.

Cross Spectral Image Reconstruction Using a Deep Guided Neural Network

Cross spectral camera arrays, where each camera records different spectral content, are becoming increasingly popular for RGB, multispectral and hyperspectral imaging, since they are capable of a high resolution in every dimension using off-the-shelf hardware. For these, it is necessary to build an image processing pipeline to calculate a consistent image data cube, i.e., it should look like as if every camera records the scene from the center camera. Since the cameras record the scene from a different angle, this pipeline needs a reconstruction component for pixels that are not visible to peripheral cameras. For that, a novel deep guided neural network (DGNet) is presented. Since only little cross spectral data is available for training, this neural network is highly regularized. Furthermore, a new data augmentation process is introduced to generate the cross spectral content. On synthetic and real multispectral camera array data, the proposed network outperforms the state of the art by up to 2 dB in terms of PSNR on average. Besides, DGNet also tops its best competitor in terms of SSIM as well as in runtime by a factor of nearly 12. Moreover, a qualitative evaluation reveals visually more appealing results for real camera array data.

End-to-End Lidar-Camera Self-Calibration for Autonomous Vehicles

Autonomous vehicles are equipped with a multi-modal sensor setup to enable the car to drive safely. The initial calibration of such perception sensors is a highly matured topic and is routinely done in an automated factory environment. However, an intriguing question arises on how to maintain the calibration quality throughout the vehicle's operating duration. Another challenge is to calibrate multiple sensors jointly to ensure no propagation of systemic errors. In this paper, we propose CaLiCa, an end-to-end deep self-calibration network which addresses the automatic calibration problem for pinhole camera and Lidar. We jointly predict the camera intrinsic parameters (focal length and distortion) as well as Lidar-Camera extrinsic parameters (rotation and translation), by regressing feature correlation between the camera image and the Lidar point cloud. The network is arranged in a Siamese-twin structure to constrain the network features learning to a mutually shared feature in both point cloud and camera (Lidar-camera constraint). Evaluation using KITTI datasets shows that we achieve 0.154 {\deg} and 0.059 m accuracy with a reprojection error of 0.028 pixel with a single-pass inference. We also provide an ablative study of how our end-to-end learning architecture offers lower terminal loss (21% decrease in rotation loss) compared to isolated calibration

Analysis of mesh-based motion compensation in wavelet lifting of dynamical 3-D+t CT data

Factorized in the lifting structure, the wavelet transform can easily be extended by arbitrary compensation methods. Thereby, the transform can be adapted to displacements in the signal without losing the ability of perfect reconstruction. This leads to an improvement of scalability. In temporal direction of dynamic medical 3-D+t volumes from Computed Tomography, displacement is mainly given by expansion and compression of tissue. We show that these smooth movements can be well compensated with a mesh-based method. We compare the properties of triangle and quadrilateral meshes. We also show that with a mesh-based compensation approach coding results are comparable to the common slice wise coding with JPEG 2000 while a scalable representation in temporal direction can be achieved.