Synchronously pumped optical parametric oscillators (OPOs) are important tools for frequency comb (FC) generation in the mid-IR spectral range, where few suitable laser gain materials exist. For degenerate OPOs, self-phase-locking to the pump FC has been demonstrated. Here, we present a phase noise study of the carrier envelope offset frequency, revealing a -6  dB reduction compared to the pump FC over a wide Fourier frequency range. These results demonstrate that a degenerate OPO can be an ideal coherent frequency divider without any excess noise.

We present an experimental and numerical study on the spectrally resolved pump-to-output intensity noise coupling in soliton fiber oscillators. In our study, we observe a strong pump noise coupling to the Kelly sidebands, while the coupling to the soliton pulse is damped. This behavior is observed in erbium-doped as well as holmium-doped fiber oscillators and confirmed by numerical modeling. It can be seen as a general feature of laser oscillators in which soliton pulse formation is dominant. We show that spectral blocking of the Kelly sidebands outside the laser cavity can improve the intensity noise performance of the laser dramatically.

Fig. 1. Our method takes a single image as the input and predicts the full and spatially-varying indoor illumination that can be used to generate consistent shading and realistic shadows after inserting virtual objects into the image.Lighting prediction from a single image is becoming increasingly important in many vision and augmented reality (AR) applications in which shading and shadow consistency between virtual and real objects should be guaranteed. However, this is a notoriously ill-posed problem, especially for indoor scenarios, because of the complexity of indoor luminaires and the limited information involved in 2D images. In this paper, we propose a graph learning-based framework for indoor lighting estimation. At its core is a new lighting model (dubbed DSGLight) based on depth-augmented Spherical Gaussians (SG) and a Graph Convolutional Network (GCN) that infers the new lighting representation from a single LDR image of limited field-ofview. Our lighting model builds 128 evenly distributed SGs over the indoor panorama, where each SG encoding the lighting and the depth around that node. The proposed GCN then learns the mapping from the input image to DSGLight. Compared with existing lighting models, our DSGLight encodes both direct lighting and indirect environmental lighting more faithfully and compactly. It also makes network training and inference more stable. The estimated depth distribution enables temporally stable shading and shadows under spatially-varying lighting. Through thorough experiments, we show that our method obviously outperforms existing methods both qualitatively and quantitatively.CCS Concepts: • Computing methodologies → Mixed / augmented reality; Scene understanding.

No abstract

No abstract