Track topics on Twitter Track topics that are important to you
Learning based depth estimation from light field has made significant progresses in recent years. However, most existing approaches are under the supervised framework, which requires vast quantities of ground-truth depth data for training. Furthermore, accurate depth maps of light field are hardly available except for a few synthetic datasets. In this paper, we exploit the multi-orientation epipolar geometry of light field and propose an unsupervised monocular depth estimation network. It predicts depth from the central view of light field without any ground-truth information. Inspired by the inherent depth cues and geometry constraints of light field, we then introduce three novel unsupervised loss functions: photometric loss, defocus loss and symmetry loss. We have evaluated our method on a public 4D light field synthetic dataset. As the first unsupervised method published in the 4D Light Field Benchmark website, our method can achieve satisfactory performance in most error metrics. Comparison experiments with two state-of-the-art unsupervised methods demonstrate the superiority of our method. We also prove the effectiveness and generality of our method on real-world light-field images.
This article was published in the following journal.
Name: IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
In this paper, we address the problem of monocular depth estimation when only a limited amount of training image-depth pairs are available. To achieve high regression accuracy, state-of-the-art estima...
Recent deep monocular depth estimation approaches based on supervised regression have achieved remarkable performance. However, they require costly ground truth annotations during training. To cope wi...
In this paper, we propose a learning based depth estimation framework suitable for both densely and sparsely sampled light fields. The proposed framework consists of three processing steps: initial de...
In this paper, we propose a novel method to generate stereoscopic images from light-field images with the intended depth range and simultaneously perform image super-resolution. Subject to the small b...
This paper proposes a unified depth model based on the light field gradient, in which estimated disparity is represented by the complex number. The complex-valued disparity by the proposed depth model...
Optimal chest compression depth during CPR is 4.56cm which is at variance with the current guidelines of 5.0-6.0cm. A change in guidelines is only worthwhile if healthcare professionals ca...
Perimetry is a well-established method that is used to measure the visual field functions of humans. Commercially available products, such as the OCTOPUS 900 (Haag-Streit AG, Koeniz, Switz...
New tools are needed to 1) diagnose and 2) stage early esophageal squamous cell carcinoma (SCC) in order to improve outcomes of this frequent and lethal cancer. Optical coherence tomograp...
Glaucoma is a leading cause of blindness worldwide. Damage to the optic nerve results in vision loss gradually. This vision loss can be detected by using visual field testing, and is a way...
The purpose of this study is to determine whether depth of anesthesia has an effect on emergence agitation (EA) in children age 2 - 8 years old. EA is a common problem in pediatric patient...
A light microscopic technique in which only a small spot is illuminated and observed at a time. An image is constructed through point-by-point scanning of the field in this manner. Light sources may be conventional or laser, and fluorescence or transmitted observations are possible.
Analytical technique for studying substances present at enzyme concentrations in single cells, in situ, by measuring light absorption. Light from a tungsten strip lamp or xenon arc dispersed by a grating monochromator illuminates the optical system of a microscope. The absorbance of light is measured (in nanometers) by comparing the difference between the image of the sample and a reference image.
The use of instrumentation and techniques for visualizing material and details that cannot be seen by the unaided eye. It is usually done by enlarging images, transmitted by light or electron beams, with optical or magnetic lenses that magnify the entire image field. With scanning microscopy, images are generated by collecting output from the specimen in a point-by-point fashion, on a magnified scale, as it is scanned by a narrow beam of light or electrons, a laser, a conductive probe, or a topographical probe.
Improvement of the quality of a picture by various techniques, including computer processing, digital filtering, echocardiographic techniques, light and ultrastructural MICROSCOPY, fluorescence spectrometry and microscopy, scintigraphy, and in vitro image processing at the molecular level.
A method of delineating blood vessels by subtracting a tissue background image from an image of tissue plus intravascular contrast material that attenuates the X-ray photons. The background image is determined from a digitized image taken a few moments before injection of the contrast material. The resulting angiogram is a high-contrast image of the vessel. This subtraction technique allows extraction of a high-intensity signal from the superimposed background information. The image is thus the result of the differential absorption of X-rays by different tissues.
Standard antiretroviral therapy (ART) consists of the combination of at least three antiretroviral (ARV) drugs to maximally suppress the HIV virus and stop the progression of HIV disease. Huge reductions have been seen in rates of death and suffering whe...