Post-capture Refocusing Research Articles

Attention Mechanism-Based Light-Field View Synthesis

The angular information of light lost in conventional images but preserved and stored in light-fields plays an instrumental role in many applications such as depth estimation, 3D reconstruction and post-capture refocusing. However, the limited angular resolution of light-field images due to the consumer hardware limitations is a major drawback in its widespread adoption. In this article, we present a novel deep learning-based light-field view synthesis method from a sparse set of input views. Our proposed method, an end-to-end trainable network, utilizes convolutional block attention modules to enhance the built-in depth image-based rendering. The proposed convolutional block attention module consists of two attention modules sequentially applied in the channel and spatial dimensions to focus the network on critical features. The proposed network architecture is a combination of three sub-networks, one for stereo feature extraction, another for disparity estimation, and the last for attention-based refinement. We present two schemes for the refinement network that perform equally well but differ in the number of parameters by 1.2% and execution time by 44%. Quantitative and qualitative results on four challenging real-world datasets show the superiority of the proposed method. Our proposed method shows considerable PSNR gains, around 1 dB compared to the state of the art and around 0.5 dB over our previous method LFVS-AM. In addition, ablation studies demonstrate the effectiveness of each module of the proposed method. Finally, the parallax edge precision-recall curve shows that our method better preserves parallax details.

Real-Time Light Field Signal Processing Using 4D/5D Linear Digital Filter FPGA Circuits

Light fields (LFs) and light field videos (LFVs) capture both angular and spatial variation of light rays emanating from scenes. This richness of information leads to novel applications such as post-capture refocusing, depth estimation and depth-velocity filtering which are not possible with images and videos. These capabilities come, however, with a significant increase in data to be processed. In order to fully exploit opportunities provided by LFs and LFVs, low-complexity signal processing algorithms that process LF and LFV data in real-time are required. In this brief, we survey such state-of-the-art algorithms, in particular for depth filtering, refocusing and denoising of LFs and depth-velcoty filtering for LFVs, and future directions for these real-time LF an LFV processing algorithms.

Multi-Volumetric Refocusing of Light Fields

Geometric information of scenes available with four-dimensional (4-D) light fields (LFs) paves the way for post-capture refocusing. Light field refocusing methods proposed so far have been limited to a single planar or a volumetric region of a scene. In this letter, we demonstrate simultaneous refocusing of multiple volumetric regions in LFs. To this end, we employ a 4-D sparse finite-extent impulse response (FIR) filter consisting of multiple hyperfan-shaped passbands. We design the 4-D sparse FIR filter as an optimal filter in the least-squares sense. Experimental results confirm that the proposed filter provides 63% average reduction in computational complexity with negligible degradation in the fidelity of multi-volumetric refocused LFs compared to a 4-D nonsparse FIR filter.

Deep Coarse-to-Fine Dense Light Field Reconstruction With Flexible Sampling and Geometry-Aware Fusion.

A densely-sampled light field (LF) is highly desirable in various applications, such as 3-D reconstruction, post-capture refocusing and virtual reality. However, it is costly to acquire such data. Although many computational methods have been proposed to reconstruct a densely-sampled LF from a sparsely-sampled one, they still suffer from either low reconstruction quality, low computational efficiency, or the restriction on the regularity of the sampling pattern. To this end, we propose a novel learning-based method, which accepts sparsely-sampled LFs with irregular structures, and produces densely-sampled LFs with arbitrary angular resolution accurately and efficiently. We also propose a simple yet effective method for optimizing the sampling pattern. Our proposed method, an end-to-end trainable network, reconstructs a densely-sampled LF in a coarse-to-fine manner. Specifically, the coarse sub-aperture image (SAI) synthesis module first explores the scene geometry from an unstructured sparsely-sampled LF and leverages it to independently synthesize novel SAIs, in which a confidence-based blending strategy is proposed to fuse the information from different input SAIs, giving an intermediate densely-sampled LF. Then, the efficient LF refinement module learns the angular relationship within the intermediate result to recover the LF parallax structure. Comprehensive experimental evaluations demonstrate the superiority of our method on both real-world and synthetic LF images when compared with state-of-the-art methods. In addition, we illustrate the benefits and advantages of the proposed approach when applied in various LF-based applications, including image-based rendering and depth estimation enhancement. The code is available at https://github.com/jingjin25/LFASR-FS-GAF.

Local Fourier Slice Photography

Light field cameras provide intriguing possibilities, such as post-capture refocus or the ability to synthesize images from novel viewpoints. This comes, however, at the price of significant storage requirements. Compression techniques can be used to reduce these, but refocusing and reconstruction require so far, again, a dense pixel representation. To avoid this, we introduce local Fourier slice photography that allows for refocused image reconstruction directly from a sparse wavelet representation of a light field, either to obtain an image or a compressed representation of it. The result is made possible by wavelets that respect the “slicing’s” intrinsic structure and enable us to derive exact reconstruction filters for the refocused image in closed-form. Image reconstruction then amounts to applying these filters to the light field’s wavelet coefficients; hence, no reconstruction of a dense pixel representation is required. We demonstrate that this can reduce storage requirements and also computation times. We furthermore analyze the computational complexity of our algorithm and show that it scales linearly with the size of the reconstructed region and the non-negligible wavelet coefficients, i.e., with the visual complexity.

Harnessing Multi-View Perspective of Light Fields for Low-Light Imaging.

Light Field (LF) offers unique advantages such as post-capture refocusing and depth estimation, but low-light conditions severely limit these capabilities. To restore low-light LFs we should harness the geometric cues present in different LF views, which is not possible using single-frame low-light enhancement techniques. We propose a deep neural network L3Fnet for Low-Light Light Field (L3F) restoration, which not only performs visual enhancement of each LF view but also preserves the epipolar geometry across views. We achieve this by adopting a two-stage architecture for L3Fnet. Stage-I looks at all the LF views to encode the LF geometry. This encoded information is then used in Stage-II to reconstruct each LF view. To facilitate learning-based techniques for low-light LF imaging, we collected a comprehensive LF dataset of various scenes. For each scene, we captured four LFs, one with near-optimal exposure and ISO settings and the others at different levels of low-light conditions varying from low to extreme low-light settings. The effectiveness of the proposed L3Fnet is supported by both visual and numerical comparisons on this dataset. To further analyze the performance of low-light restoration methods, we also propose the L3F-wild dataset that contains LF captured late at night with almost zero lux values. No ground truth is available in this dataset. To perform well on the L3F-wild dataset, any method must adapt to the light level of the captured scene. To do this we use a pre-processing block that makes L3Fnet robust to various degrees of low-light conditions. Lastly, we show that L3Fnet can also be used for low-light enhancement of single-frame images, despite it being engineered for LF data. We do so by converting the single-frame DSLR image into a form suitable to L3Fnet, which we call as pseudo-LF. Our code and dataset is available for download at https://mohitlamba94.github.io/L3Fnet/.

Dual Disparity-Based Novel View Reconstruction for Light Field Images Using Discrete Cosine Transform Filter

Compared to conventional photography which allows capturing spatial intensity only, light field imaging can capture angular and spatial information by collecting information from all directions. This massive information could be used in many applications such as depth estimation and post-capture refocusing. In addition, the emergence of consumer light field cameras has resulted in the widespread use of light field imaging. However, its limited resolution constitutes a major drawback to the use of enormous capabilities provided. In this article, we tried to alleviate the effect of this drawback by using a machine learning algorithm. Our proposed network is built upon existing reconstruction techniques that divide the process into disparity estimation and final image reconstruction. This is achieved by using two consecutive neural networks while the whole network is trained simultaneously. We propose to use a predefined convolutional network at the first stage to decrease preprocessing time, and in addition, we use dual disparity vectors to alleviate the interpolation error whereas warping input images. Our system was trained to reconstruct multi-angular resolution images fast and accurately using real light field images for training. Experimental results demonstrate that the proposed system can reconstruct high-quality images faster than the state-of-the-art techniques.

Light Field Reconstruction With a Hybrid Sparse Regularization-Pseudo 4DCNN Framework

The densely-sampled light field (LF) is highly desirable in many applications, such as 3D reconstruction, post-capture refocusing and virtual/augmented reality. However, it is costly and challenging to capture them because of the high dimensional nature. Existing view synthesis methods employ depth information for the densely-sampled LF reconstruction from the undersampled LF with a large disparity range, but fail to achieve non-Lambertian performance. In this article, a novel coarse-to-fine LF reconstruction method is proposed. We will develop a hybrid reconstruction framework that fuses model-based sparse regularization with deep learning. Specifically, sparse regularization with directional filter bank is utilized to solve the large baseline problem and gives a coarse densely-sampled LF. In addition, for those that cannot be recovered by classical model-based methods due to limited angular resolution, is estimated by learning a pseudo 4D convolutional neural network, and thereby offering a refinement on the intermediate LF. Extensive experiments demonstrate the superiority of our method on both real-world and synthetic LF images when compared with state-of-the-art methods.

Light-field image acquisition from a conventional camera: design of a four minilens ring device

In the past few years, a new type of camera has been emerging on the market: a digital camera capable of capturing both the intensity of the light emanating from a scene and the direction of the light rays. This camera technology called a light-field camera uses an array of lenses placed in front of a single image sensor, or simply, an array of cameras attached together. An optical device is proposed: a four minilens ring that is inserted between the lens and the image sensor of a digital camera. This device prototype is able to convert a regular digital camera into a light-field camera as it makes it possible to record four subaperture images of the scene. It is a compact and cost-effective solution to perform both postcapture refocusing and depth estimation. The minilens ring makes also the plenoptic camera versatile; it is possible to adjust the parameters of the ring so as to reduce or increase the size of the projected image. Together with the proof of concept of this device, we propose a method to estimate the positions of each optical component depending on the observed scene (object size and distance) and the optics parameters. Real-world results are presented to validate our device prototype.

Geometry-Consistent Light Field Super-Resolution via Graph-Based Regularization.

Light field cameras capture the 3D information in a scene with a single exposure. This special feature makes light field cameras very appealing for a variety of applications: from post-capture refocus to depth estimation and image-based rendering. However, light field cameras suffer by design from strong limitations in their spatial resolution. Off-the-shelf super-resolution algorithms are not ideal for light field data, as they do not consider its structure. On the other hand, the few super-resolution algorithms explicitly tailored for light field data exhibit significant limitations, such as the need to carry out a costly disparity estimation procedure with sub-pixel precision. We propose a new light field super-resolution algorithm meant to address these limitations. We use the complementary information in the different light field views to augment the spatial resolution of the whole light field at once. In particular, we show that coupling the multi-view approach with a graph-based regularizer, which enforces the light field geometric structure, permits to avoid the need of a precise and costly disparity estimation step. Extensive experiments show that the new algorithm compares favorably to the state-of-the-art methods for light field super-resolution, both in terms of visual quality and in terms of reconstruction error.

Light field super resolution through controlled micro-shifts of light field sensor

Light field cameras enable new capabilities, such as post-capture refocusing and aperture control, through capturing directional and spatial distribution of light rays in space. Micro-lens array based light field camera design is often preferred due to its light transmission efficiency, cost-effectiveness and compactness. One drawback of the micro-lens array based light field cameras is low spatial resolution due to the fact that a single sensor is shared to capture both spatial and angular information. To address the low spatial resolution issue, we present a light field imaging approach, where multiple light fields are captured and fused to improve the spatial resolution. For each capture, the light field sensor is shifted by a pre-determined fraction of a micro-lens size using an xy translation stage for optimal performance.

Spatial and Angular Resolution Enhancement of Light Fields Using Convolutional Neural Networks.

Light field imaging extends the traditional photography by capturing both spatial and angular distribution of light, which enables new capabilities, including post-capture refocusing, post-capture aperture control, and depth estimation from a single shot. Micro-lens array (MLA) based light field cameras offer a cost-effective approach to capture light field. A major drawback of MLA based light field cameras is low spatial resolution, which is due to the fact that a single image sensor is shared to capture both spatial and angular information. In this paper, we present a learning based light field enhancement approach. Both spatial and angular resolution of captured light field is enhanced using convolutional neural networks. The proposed method is tested with real light field data captured with a Lytro light field camera, clearly demonstrating spatial and angular resolution improvement.

Light Field Image Processing: An Overview

Light field imaging has emerged as a technology allowing to capture richer visual information from our world. As opposed to traditional photography, which captures a 2D projection of the light in the scene integrating the angular domain, light fields collect radiance from rays in all directions, demultiplexing the angular information lost in conventional photography. On the one hand, this higher dimensional representation of visual data offers powerful capabilities for scene understanding, and substantially improves the performance of traditional computer vision problems such as depth sensing, post-capture refocusing, segmentation, video stabilization, material classification, etc. On the other hand, the high-dimensionality of light fields also brings up new challenges in terms of data capture, data compression, content editing, and display. Taking these two elements together, research in light field image processing has become increasingly popular in the computer vision, computer graphics, and signal processing communities. In this paper, we present a comprehensive overview and discussion of research in this field over the past 20 years. We focus on all aspects of light field image processing, including basic light field representation and theory, acquisition, super-resolution, depth estimation, compression, editing, processing algorithms for light field display, and computer vision applications of light field data.

Camera array based light field microscopy.

This paper proposes a novel approach for high-resolution light field microscopy imaging by using a camera array. In this approach, we apply a two-stage relay system for expanding the aperture plane of the microscope into the size of an imaging lens array, and utilize a sensor array for acquiring different sub-apertures images formed by corresponding imaging lenses. By combining the rectified and synchronized images from 5 × 5 viewpoints with our prototype system, we successfully recovered color light field videos for various fast-moving microscopic specimens with a spatial resolution of 0.79 megapixels at 30 frames per second, corresponding to an unprecedented data throughput of 562.5 MB/s for light field microscopy. We also demonstrated the use of the reported platform for different applications, including post-capture refocusing, phase reconstruction, 3D imaging, and optical metrology.

Generalized assorted camera arrays: robust cross-channel registration and applications.

One popular technique for multimodal imaging is generalized assorted pixels (GAP), where an assorted pixel array on the image sensor allows for multimodal capture. Unfortunately, GAP is limited in its applicability because of the need for multimodal filters that are amenable with semiconductor fabrication processes and results in a fixed multimodal imaging configuration. In this paper, we advocate for generalized assorted camera (GAC) arrays for multimodal imaging--i.e., a camera array with filters of different characteristics placed in front of each camera aperture. The GAC provides us with three distinct advantages over GAP: ease of implementation, flexible application-dependent imaging since filters are external and can be changed and depth information that can be used for enabling novel applications (e.g., postcapture refocusing). The primary challenge in GAC arrays is that since the different modalities are obtained from different viewpoints, there is a need for accurate and efficient cross-channel registration. Traditional approaches such as sum-of-squared differences, sum-of-absolute differences, and mutual information all result in multimodal registration errors. Here, we propose a robust cross-channel matching cost function, based on aligning normalized gradients, which allows us to compute cross-channel subpixel correspondences for scenes exhibiting nontrivial geometry. We highlight the promise of GAC arrays with our cross-channel normalized gradient cost for several applications such as low-light imaging, postcapture refocusing, skin perfusion imaging using color + near infrared, and hyperspectral imaging.

PiCam

We presentPiCam(Pelican Imaging Camera-Array), an ultra-thin high performance monolithic camera array, that captures light fields and synthesizes high resolution images along with a range image (scene depth) through integrated parallax detection and superresolution. The camera is passive, supporting both stills and video, low light capable, and small enough to be included in the next generation of mobile devices including smartphones. Prior works [Rander et al. 1997; Yang et al. 2002; Zhang and Chen 2004; Tanida et al. 2001; Tanida et al. 2003; Duparré et al. 2004] in camera arrays have explored multiple facets of light field capture - from viewpoint synthesis, synthetic refocus, computing range images, high speed video, and micro-optical aspects of system miniaturization. However, none of these have addressed the modifications needed to achieve the strict form factor and image quality required to make array cameras practical for mobile devices. In our approach, we customize many aspects of the camera array including lenses, pixels, sensors, and software algorithms to achieve imaging performance and form factor comparable to existing mobile phone cameras.Our contributions to the post-processing of images from camera arrays include a cost function for parallax detection that integrates across multiple color channels, and a regularized image restoration (superresolution) process that takes into account all the system degradations and adapts to a range of practical imaging conditions. The registration uncertainty from the parallax detection process is integrated into a Maximum-a-Posteriori formulation that synthesizes an estimate of the high resolution image and scene depth. We conclude with some examples of our array capabilities such as postcapture (still) refocus, video refocus, view synthesis to demonstrate motion parallax, 3D range images, and briefly address future work.

A Light-Field Image Sensor in 180 nm CMOS

This paper presents a CMOS image sensor which captures local incident angle and intensity information from the light it records. The 400×384 pixel array employs 7.5 μm angle-sensitive pixels, which use pairs of local diffraction gratings above a photodiode to detect incident angle. The gratings are implemented with the metal interconnect layers of CMOS manufacturing technology and therefore require no post-processing or external optics. Fabricated in a 180 nm mixed-mode CMOS process, the sensor requires only a single lens to create a light-field image. Using the information contained in a single image, we demonstrate range finding with 2.5 mm precision at 1 m and post-capture refocus on complex visual scenes.