Lightweight Deep Exemplar Colorization via Semantic Attention-Guided Laplacian Pyramid.

Exemplar-based colorization aims to generate plausible colors for a grayscale image with the guidance of a color reference image. The main challenging problem is finding the correct semantic correspondence between the target image and the reference image. However, the colors of the object and background are often confused in the existing methods. Besides, these methods usually use simple encoder-decoder architectures or pyramid structures to extract features and lack appropriate fusion mechanisms, which results in the loss of high-frequency information or high complexity. To address these problems, this paper proposes a lightweight semantic attention-guided Laplacian pyramid network (SAGLP-Net) for deep exemplar-based colorization, exploiting the inherent multi-scale properties of color representations. They are exploited through a Laplacian pyramid, and semantic information is introduced as high-level guidance to align the object and background information. Specially, a semantic guided non-local attention fusion module is designed to exploit the long-range dependency and fuse the local and global features. Moreover, a Laplacian pyramid fusion module based on criss-cross attention is proposed to fuse high frequency components in the large-scale domain. An unsupervised multi-scale multi-loss training strategy is further introduced for network training, which combines pixel loss, color histogram loss, total variance regularisation, and adversarial loss. Experimental results demonstrate that our colorization method achieves better subjective and objective performance with lower complexity than the state-of-the-art methods.

Octave soliton microcombs in lithium niobate microresonators.

Soliton microcombs are regarded as an ideal platform for applications such as optical communications, optical sensing, low-noise microwave sources, optical atomic clocks, and frequency synthesizers. Many of these applications require a broad comb spectrum that covers an octave, essential for implementing the f - 2f self-referencing techniques. In this work, we have successfully generated an octave-spanning soliton microcomb based on a z-cut thin-film lithium niobate (TFLN) microresonator. This achievement is realized under on-chip optical pumping at 340 mW and through extensive research into the broadening of dual dispersive waves (DWs). Furthermore, the repetition rate of the octave soliton microcomb is accurately measured using an electro-optic comb generated by an x-cut TFLN racetrack microresonator. Our results represent a crucial step toward the realization of practical, integrated, and fully stabilized soliton microcomb systems based on TFLN.

On-Chip Meta-Optics for Engineering Arbitrary Trajectories with Longitudinal Polarization Variation.

On-chip integrated meta-optics promise to achieve high-performance and compact integrated photonic devices. To arbitrarily engineer the optical trajectory along the propagation path in an on-chip integrated scheme is of significance in fundamental physics and various emerging applications. Here, we experimentally demonstrate an on-chip metasurface integrated on a waveguide to enable predefined arbitrary optical trajectories in the visible regime. By transformation of the transverse phase to generate longitudinal mapping, the guided waves are extracted and molded into any different optical trajectories (parabola, hyperbola, and cosine). More intriguingly, predefined polarization states with longitudinal variation are also successfully imparted along the trajectory. Owing to the on-chip propagation scheme, the trajectories are uniquely free from zero-order diffraction interference, naturally having a higher signal-to-noise ratio beyond conventional free-space forms. Overall, such on-chip optical trajectory engineering allows for miniaturized integration and can find paths in potential applications of complex optical manipulation, advanced laser fabrication, and microscopic imaging.

Repetition rate tuning and locking of solitons in a microrod resonator.

Recently, there has been significant interest in the generation of coherent temporal solitons in optical microresonators. In this Letter, we present a demonstration of dissipative Kerr soliton generation in a microrod resonator using an auxiliary-laser-assisted thermal response control method. In addition, we are able to control the repetition rate of the soliton over a range of 200 kHz while maintaining the pump laser frequency, by applying external stress tuning. Through the precise control of the PZT voltage, we achieve a stability level of 3.9 × 10-10 for residual fluctuation of the repetition rate when averaged 1 s. Our platform offers precise tuning and locking capabilities for the repetition frequency of coherent mode-locked combs in microresonators. This advancement holds great potential for applications in spectroscopy and precision measurements.

Switchable unidirectional emissions from hydrogel gratings with integrated carbon quantum dots.

Directional emission of photoluminescence despite its incoherence is an attractive technique for light-emitting fields and nanophotonics. Optical metasurfaces provide a promising route for wavefront engineering at the subwavelength scale, enabling the feasibility of unidirectional emission. However, current directional emission strategies are mostly based on static metasurfaces, and it remains a challenge to achieve unidirectional emissions tuning with high performance. Here, we demonstrate quantum dots-hydrogel integrated gratings for actively switchable unidirectional emission with simultaneously a narrow divergence angle less than 1.5° and a large diffraction angle greater than 45°. We further demonstrate that the grating efficiency alteration leads to a more than 7-fold tuning of emission intensity at diffraction order due to the variation of hydrogel morphology subject to change in ambient humidity. Our proposed switchable emission strategy can promote technologies of active light-emitting devices for radiation control and optical imaging.

Multi-Dimensional Light-Emitting Meta-Display: Photoluminescence and Pumping Light Multiplexing.

The advent of intelligent display devices has given rise to diverse and complex demands for miniature light-emitting devices. Light-emitting metasurfaces have emerged as a practical and efficient means of achieving precise light modulation. However, their practicality is limited by certain constraints. First, there is a need for further exploration of the ability to manipulate both pumping and emitting light simultaneously. Second, there is currently no encoding freedom in multi-dimensional emitting light. To address these concerns, using meta-atoms is proposed to encode both fluorescence and pumping light independently, and expand the encoding freedom with different incident wavevector directions. A light-emitting metasurface with quad-fold multiplex encoding meta-displays, including dual scattering images and dual fluorescence images, is further demonstrated. This design strategy not only manipulates both pumping and fluorescence light but also broadens encoding freedom for comprehensive multi-functionality. This can pave the way for multiplexing optical displays, information storage, and next-generation wearable displays.

Adaptive Chroma Prediction Based on Luma Difference for H.266/VVC.

Cross-component chroma prediction plays an important role in improving coding efficiency for H.266/VVC. We use the differences between reference samples and the predicted sample to design an attention model for chroma prediction, namely luma difference-based chroma prediction (LDCP). Specifically, the luma differences (LDs) between reference samples and the predicted sample are employed as the input of the attention model, which is designed as a softmax function to map LDs to chroma weights nonlinearly. Finally, a weighted chroma prediction is conducted based on the weights and chroma reference samples. To provide adaptive weights, the model parameter of the softmax function can be determined based on the template (T-LDCP) or offline learning (L-LDCP), respectively. Experimental results show that the T-LDCP achieves BD-rate reductions of 0.34%, 2.02%, and 2.34% for the Y, Cb, and Cr components, and the L-LDCP brings 0.32%, 2.06%, and 2.21% BD-rate savings for Y, Cb, and Cr components, respectively. The L-LDCP introduces slight encoding and decoding time increments, i.e., 2% and 1%, when integrated into the latest VVC test model version 18.0. Besides, the LDCP can be implemented by a pixel-level parallelization which is hardware-friendly.

Short-term association of CO and NO2 with hospital visits for glomerulonephritis in Hefei, China: a time series study.

Objective: Recent studies suggest air pollution as an underlying factor to kidney disease. However, there is still limited knowledge about the short-term correlation between glomerulonephritis (GN) and air pollution. Thus, we aim to fill this research gap by investigating the short-term correlation between GN clinical visits and air pollution exposure. Methods: Between 2015 and 2019, daily GN visit data from two grade A tertiary hospitals in Hefei City were collected, along with corresponding air pollution and meteorological data. A generalized linear model integrated with a distributed lag nonlinear model was employed to analyze the relationship between GN visits and air pollutants. Moreover, we incorporated a dual pollutant model to account for the combined effects of multiple pollutants. Furthermore, subgroup analyses were performed to identify vulnerable populations based on gender, age, and season. Results: The association between 23,475 GN visits and air pollutants was assessed, and significant positive associations were found between CO and NO2 exposure and GN visit risk. The single-day lagged effect model for CO showed increased risks for GN visits from lag0 (RR: 1.129, 95% CI: 1.031-1.236) to lag2 (RR: 1.034, 95% CI: 1.011-1.022), with the highest risk at lag0. In contrast, NO2 displayed a more persistent impact (lag1-lag4) on GN visit risk, peaking at lag2 (RR: 1.017, 95% CI: 1.011-1.022). Within the dual-pollutant model, the significance persisted for both CO and NO2 after adjusting for each other. Subgroup analyses showed that the cumulative harm of CO was greater in the cold-season and older adult groups. Meanwhile, the female group was more vulnerable to the harmful effects of cumulative exposure to NO2. Conclusion: Our study indicated that CO and NO2 exposure can raise the risk of GN visits, and female and older adult populations exhibited greater susceptibility.

Thermal oscillation in the hybrid Si3N4 - TiO2 microring.

The hybrid microcavity composed of different materials shows unique thermal-optical properties such as resonance frequency shift and small thermal noise fluctuations with the temperature variation. Here, we have fabricated the hybrid Si3N4 - TiO2 microring, which decreases the effective thermo-optical coefficients (TOC) from 23.2pm/K to 11.05pm/K due to the opposite TOC of these two materials. In this hybrid microring, we experimentally study the thermal dynamic with different input powers and scanning speeds. The distorted transmission and thermal oscillation are observed, which results from the non-uniform scanning speed and the different thermal relaxation times of the Si3N4 and the TiO2. We calibrate the distorted transmission spectrum for the resonance measurement at the reverse scanning direction and explain the thermal oscillation with a thermal-optical coupled model. Finally, we analyse the thermal oscillation condition and give the diagram about the oscillation region, which has significant guidance for the occurrence and avoidance of the thermal oscillation in practical applications.

kHz-precision wavemeter based on reconfigurable microsoliton.

The mode-locked microcomb offers a unique and compact solution for photonics applications, ranging from the optical communications, the optical clock, optical ranging, the precision spectroscopy, novel quantum light source, to photonic artificial intelligence. However, the photonic micro-structures are suffering from the perturbations arising from environment thermal noises and also laser-induced nonlinear effects, leading to the frequency instability of the generated comb. Here, a universal mechanism for fully stabilizing the microcomb is proposed and experimentally verified. By incorporating two global tuning approaches and the autonomous thermal locking mechanism, the pump laser frequency and repetition rate of the microcomb can be controlled independently in real-time without interrupting the microcomb generation. The high stability and controllability of the microcomb frequency enables its application in wavelength measurement with a precision of about 1 kHz. The approach for the full control of comb frequency could be applied in various microcomb platforms, and improve their performances in timing, spectroscopy, and sensing.

Toward Water-Immersion Programmable Meta-Display.

Heading toward next-generation intelligent display, dynamic control capability for meta-devices is critical for real world applications. Beyond the conventional electrical/optical/mechanical/thermal tuning methods, liquid immersion recently has emerged as a facile tuning mechanism which is easily accessible (especially water) and practically implementable for large tuning area. However, due to the longstanding and critical drawback of lacking independent-encoding capability, the state-of-art immersion approach remains incapable of pixel-level programmable switching. Here a water-immersion tuning scheme with pixel-scale programmability for dynamic meta-displays is proposed. Tunable meta-pixels can be engineered to construct spectral selective patterns at prior-/post- immersion states, such that a metasurface enables pixel-level transforming animations for dynamic multifield meta-displays, including near-field dual-nanoprints and far-field dual-holographic displays. The proposed water-immersion programmable approach for meta-display, benefitting from its large tuning area, facile operation and strong repeatability, may find a revolutionary path toward next-generation intelligent display with practical applications in dynamic display/encryption, information anticounterfeit/storage, and optical sensors.

Numerical characterization of soliton microcomb in an athermal hybrid Si3N4-TiO2 microring.

We theoretically investigate the athermal constructions to cancel the thermorefractive effect of a hybrid Si3N4-TiO2 microring, which merges two materials with opposite thermo-optical coefficients (TOCs). The analytical and numerical results predict that the thermorefractive effect can be reduced under the appropriate parameters. In addition, the soliton state is easily accessed under the athermal condition. The thermorefractive noise due to the fluctuation of the microresonator temperature caused by the heat exchange between the microresonator and the surrounding environment is also suppressed by one order of magnitude, which is critical for the potential applications of soliton microcombs, such as spectroscopy, optical clocks and microwave generation.

All-Optical Synchronization of Remote Optomechanical Systems.

Synchronization and frequency locking between remote mechanical oscillators are of scientific and technological importance. The key challenges are to align the oscillation frequencies and realize strong nonlinear interaction of both oscillators to a common carrier capable of long-distance transmission. Here, we experimentally realize the all-optical synchronization between two different optomechanical systems, a microsphere and a microdisk. The mechanical oscillation of the microsphere induced by the radiation pressure is loaded onto the pump laser via the optomechanical interaction, which is directly transmitted through a 5-km-long single-mode fiber to excite the mechanical oscillation of the microdisk. By finely tuning both the optical and mechanical frequencies of the two microresonators, the oscillation of the microdisk is injection locked to the microsphere, resulting in a synchronized phase relation of the two systems. Our results push a step forward the long-distance synchronization network using optomechanical microresonators.

Thermal tuning of mode crossing and the perfect soliton crystal in a Si3N4 microresonator.

Dissipative Kerr solitons in high quality microresonators have attracted much attention in the past few years. They provide ideal platforms for a number of applications. Here, we fabricate the Si3N4 microring resonator with anomalous dispersion for the generation of single soliton and soliton crystal. Based on the strong thermal effect in the high-Q microresonator, the location and strength of the avoided mode crossing in the device can be changed by the intracavity power. Because the existence of the avoided mode crossing can induce the perfect soliton crystal with specific soliton number, we could choose the appropriate pumped resonance mode and appropriate pump power to obtain the perfect soliton crystals on demand.

Graph Convolutional Dictionary Selection With L2,ₚ Norm for Video Summarization.

Video Summarization (VS) has become one of the most effective solutions for quickly understanding a large volume of video data. Dictionary selection with self representation and sparse regularization has demonstrated its promise for VS by formulating the VS problem as a sparse selection task on video frames. However, existing dictionary selection models are generally designed only for data reconstruction, which results in the neglect of the inherent structured information among video frames. In addition, the sparsity commonly constrained by L2,1 norm is not strong enough, which causes the redundancy of keyframes, i.e., similar keyframes are selected. Therefore, to address these two issues, in this paper we propose a general framework called graph convolutional dictionary selection with L2,p ( ) norm (GCDS 2,p ) for both keyframe selection and skimming based summarization. Firstly, we incorporate graph embedding into dictionary selection to generate the graph embedding dictionary, which can take the structured information depicted in videos into account. Secondly, we propose to use L2,p ( ) norm constrained row sparsity, in which p can be flexibly set for two forms of video summarization. For keyframe selection, can be utilized to select diverse and representative keyframes; and for skimming, p=1 can be utilized to select key shots. In addition, an efficient iterative algorithm is devised to optimize the proposed model, and the convergence is theoretically proved. Experimental results including both keyframe selection and skimming based summarization on four benchmark datasets demonstrate the effectiveness and superiority of the proposed method.

Dual-encryption freedom via a monolayer-nanotextured Janus metasurface in the broadband visible.

As an emerging category of two-faced 2D architecture, the Janus metasurface aims to explore another universal optical property, that is, the wavevector direction (k-direction), and to enable the asymmetric transmission between the opposite directional incidences. It exhibits significant potential in creating versatile multiplexing metasurfaces and an optical isolator in optical communication applications. However, most previous asymmetric functionality shows merely one-way functionality with the other-way simply muted or demands multilayered nanostructure fabrication and alignment. Hence, it remains a great challenge to make a monolayer-nanotextured Janus metasurface with dual-encryption freedom and conquering the difficulty for multilayer alignment and practical operation bandwidth. In this work, we have proposed and experimentally demonstrated a new strategy of a dual-encryption Janus metasurface design with a simple monolayer-nanotextured metasurface coupled with a commercialized film of the half-wave plate. Utilizing the hybridization from two independent geometrical dimensions of rectangular-antennas, our approach ingeniously transforms the polarization-multiplexing into the dual-directional channels. A series of calculations and experimental results demonstrate that our asymmetric approach simultaneously constructs completely independent imaging encryptions for both forward and backward directions. Additionally, our proposed approach becomes a practical scheme with broadband visible-frequency operation and great simplicity in design and nanofabrication. We believe the universal scheme could facilitate to increase the information encoding capacity and holographic multiplexing channels by expanding the illumination wavevector to the full-space (+/-), and it paves the route toward the potential applications in on-chip integration, telecommunications, encryption, information processing, and communication.

Angular Multiplexing Nanoprinting with Independent Amplitude Encryption Based on Visible-Frequency Metasurfaces.

Two-dimensional (2D) metasurfaces hold great promise to enable multiplexing and multifunctional optical devices due to their artificial freedom in design, device miniaturization, etc. Various multiplexing and multifunctional metasurfaces have been extensively studied, including polarization multiplexing, wavelength multiplexing, and orbit angular momentum (OAM) multiplexing. However, due to the lack of angular encoding freedom, angular multiplexing switchable nanoprinting has rarely been studied or demonstrated yet to the best of our knowledge. Here, we realize angular multiplexing switchable nanoprinting functionality with independent amplitude encryption based on visible-frequency metasurfaces. By screening a large number of structural designs and breaking the angular correlation, we eventually obtain optimal metasurface designs to realize dual-channel arbitrary image encryption. Furthermore, we illustrate that the proposed scheme would serve as an optical information concealment/retrieval strategy by combining the structural color and amplitude modulation. Overall, we believe that angular multiplexing metasurfaces would easily find promising applications, including optical information encryption/concealment, multifunctional switchable devices, and advanced eyeglass-free three-dimensional (3D) stereoscopic displays.

3D Meta-Prisms for Versatile Beam Steering by Hybridizing Plasmonic and Diffractive Effect in the Broadband Visible Regime.

As two independent optical sub-fields, diffraction optics and plasmonics both have been used for wavefront shaping and beam steering. However, the two separate concepts have always been developing as two parallel directions, which have not met for studying their structural hybridization to discover new potentials. For instance of the flat metasurfaces, even though the geometric parameters including shape, size, and periodicity have been studied, it remains mostly unexplored for the 3D spatial height variation. Here, a new type of all-metallic 3D meta-prism is proposed and experimentally demonstrated by hybridizing the localized surface plasmonic resonances (LSPR) and the blazed grating diffraction, which enables strong polarization-dependent behaviors to steer broadband visible light to drastically inverse directions. The nanofabrication of 3D meta-prism is achieved by nanostencil lithography with electron-beam evaporation. Such meta-prism could also enable to split different visible light (green, blue, and red) with high-efficiency contrast (≈10). By the mirror-symmetry arrangement, a multifunctional surface is demonstrated with polarization-/wavelength-multiplexing wavefront-shaping functions (concave, convex, or flat mirror). This unique 3D meta-prism enjoys great simplicity and versatility in broadband beam steering through the incorporation of plasmonic and diffractive effects and can be utilized in various applications including dichroic-prism splitters, multifunctional meta-mirrors, etc.

Observation of chiral edge states in gapped nanomechanical graphene.

Emerging in diverse areas of physics, edge states have been exploited as an efficient strategy of manipulating electrons, photons, and phonons for next-generation hybrid electro-optomechanical circuits. Among various edge states, gapless chiral edge states harnessing quantum spin/valley Hall effects in graphene or graphene-like materials are especially unique. Here, we report on an experimental demonstration of chiral edge states in gapped "nanomechanical graphene"-a honeycomb lattice of free-standing silicon nitride nanomechanical membranes with broken spatial inversion symmetry. These chiral edge states can emerge from the conventional flat-band edge states by tuning the on-site boundary potentials. We experimentally demonstrated that they are backscattering-immune against sharp bends and exhibit the "valley-momentum locking" effect. We further realized smooth transition between the chiral edge states and the well-known valley kink states. Our results open the door to experimental investigation of exotic graphene-related physics in the very-high-frequency integrated nanomechanical systems.

Keyframe Extraction From Laparoscopic Videos via Diverse and Weighted Dictionary Selection.

Laparoscopic videos have been increasingly acquired for various purposes including surgical training and quality assurance, due to the wide adoption of laparoscopy in minimally invasive surgeries. However, it is very time consuming to view a large amount of laparoscopic videos, which prevents the values of laparoscopic video archives from being well exploited. In this paper, a dictionary selection based video summarization method is proposed to effectively extract keyframes for fast access of laparoscopic videos. Firstly, unlike the low-level feature used in most existing summarization methods, deep features are extracted from a convolutional neural network to effectively represent video frames. Secondly, based on such a deep representation, laparoscopic video summarization is formulated as a diverse and weighted dictionary selection model, in which image quality is taken into account to select high quality keyframes, and a diversity regularization term is added to reduce redundancy among the selected keyframes. Finally, an iterative algorithm with a rapid convergence rate is designed for model optimization, and the convergence of the proposed method is also analyzed. Experimental results on a recently released laparoscopic dataset demonstrate the clear superiority of the proposed methods. The proposed method can facilitate the access of key information in surgeries, training of junior clinicians, explanations to patients, and archive of case files.