We report a new class of turn-on surface enhanced Raman scattering (SERS) sensors for the sensitive and selective detection cadmium ion (Cd2+) by taking advantage interparticle plasmonic coupling generated in process Cd2+-selective nanoparticle self-aggregation. The SERS-active nanoparticles consist 41-nm gold nanoparticles, encoded with Raman-active dye through disulfide anchoring group, layer Cd2+-chelating polymer brush coating grafted on via surface-initiated atom transfer radical...

Existing underwater image restoration (UIR) methods generally only handle color distortion or jointly address and haze issues, but they often overlook the more complex degradations that can occur in scenes. To this limitation, we propose a Universal Underwater Image Restoration method, termed as UniUIR, considering scenario of real-world mixed distortions an all-in-one manner. decouple degradation-specific issues explore inter-correlations among various UIR task, designed Mamba...

Distance estimation using a monocular camera is one of the most classic tasks for computer vision. Current distance estimating methods need lot data collection or they produce imprecise results. In this paper, we propose network both object detection and estimation. A network-based on ShuffleNet YOLO used to detect an object, self-supervised learning estimate distance. We calibrated camera, parameters were integrated into overall network. also analyzed parameter variation pose. Further,...

Metal oxides are of paramount importance in heterogeneous catalysis as either supports or active phases. Controlled synthesis one-dimensional (1D) metal oxide nanostructures has received enormous attention due to the possibility tailoring properties by tuning their shapes, sizes, and compositions. This feature article highlights recent advances shape controlled 1D applications catalysis, with aim introducing new insights into catalyst design.

Many 4D light field processing applications rely on superpixel segmentations, for which occlusion-aware view consistency is important. Yet, existing methods often enforce by propagating clusters from a central only, can lead to inconsistent superpixels non-central views. Our proposed approach combines an angular segmentation in horizontal and vertical EPI spaces with clustering propagation step across all Qualitative video demonstrations show that this helps remove flickering boundary shapes...

In this work, we propose a novel deep online correction (DOC) framework for monocular visual odometry. The whole pipeline has two stages: First, depth maps and initial poses are obtained from convolutional neural networks (CNNs) trained in self-supervised manners. Second, the predicted by CNNs further improved minimizing photometric errors via gradient updates of during inference phases. benefits our proposed method twofold: 1) Different online-learning methods, DOC does not need to...

Corrective lenses introduce distortion caused by the refraction effect, which changes wear's appearance. To give users a more realistic experience, virtual try-on system for prescription eyeglasses modifies an input video and virtually inserts eyeglasses, producing output similar to mirror. The proposed generates 3D representation of corrective mounted into frame sequence insert through image-based rendering. Unlike existing systems, this approach simulates effects due lens takes account...

This paper addresses active lighting recurrence (ALR), a new problem that actively relocalizes light source to physically reproduce the condition for same scene from single reference image. ALR is of great importance fine-grained visual monitoring and change detection, because some phenomena or minute changes can only be clearly observed under particular conditions. Hence, effective should able online navigate toward target pose, which challenging due complexity diversity real-world \&...

Dynamic relocalization of 6D camera pose from single reference image is a costly and challenging task that requires delicate hand-eye calibration precision positioning platform to do 3D mechanical rotation translation. In this paper, we show high-quality can be achieved in much less expensive way. Based on inexpensive with unreliable absolute repositioning accuracy (ARA), propose free strategy actively relocate into the same produces input image, by sequentially correcting relative We...

Environment depth estimation by fusing camera and radar enables a broad spectrum of applications such as autonomous driving, environmental perception, context-aware localization navigation. Various pioneering approaches have been proposed to achieve accurate dense integrating vision LiDAR through deep learning. However, due the challenges sparse sampling in-vehicle LiDARs, high ground-truth annotation overhead, severe dynamics in real environments, existing solutions not yet achieved...

Purpose Light field images (LFIs) have gained popularity as a technology to increase the of view (FoV) plenoptic cameras since they can capture information about light rays with large FoV. Wide FoV causes (LF) data rapidly, which restricts use LF imaging in image processing, visual analysis and user interface. Effective LFI coding methods become paramount importance. This paper aims eliminate more redundancy by exploring sparsity correlation angular domain LFIs, well mitigate loss perceptual...

Light field (LF) cameras can capture the intensity and angle information of any scene in one single shot, are widely used virtual reality, refocusing, de-occlusion, depth estimation, etc. However, fundamental limitation between angular spatial resolution leads to low LF images, which limits their development prospects various fields. In this paper, we propose a new super-resolution network based on view interaction hierarchical feature fusion effectively improve image preserve consistency...

We disclose a method of fabricating low-voltage-driven smart glass based on micro-patterned liquid crystal (LC) Fresnel lenses and implement three proof-of-concept prototypes. Distinct from the conventional LC-based windows with scattering state, prominence our proposed LC in blurry state under both normal oblique observations stems image distortion caused by lenses. In addition, high transmittance (>90%) clear is obtained applying low voltage 2 V to each prototype. Moreover, elaborating...

In order to get raw images of high quality for downstream Image Signal Process (ISP), in this paper we present an Efficient Locally Multiplicative Transformer called ELMformer image restoration. contains two core designs especially whose primitive attribute is single-channel. The first design a Bi-directional Fusion Projection (BFP) module, where consider both the color characteristics and spatial structure second one that propose Self-Attention (L-MSA) scheme effectively deliver information...

In our turbine performance assessment, we need to provide an effective visual analytics tool in handling high-dimensional datasets. We have employed RadViz 2D exploratory data analysis. However, with the increase of dataset size and dimensionality, clumping projected points towards origin causes low space utilization, which largely degenerates visibility feature characteristics. this study, better evaluate hidden patterns center region, propose histogram-based techniques manipulate radial...

In this paper, an accurate off-line map-matching (OM2) system is designed for complex trajectory networks. It difficult to trajectories input into the hidden Markov model (HMM) directly. OM2 includes three key modules that are pre-processing, based on weight adaptation HMM (WA-HMM), and post-processing. The pre-processing module divides multi-trajectory single-trajectory sets self-defined division (TDM) of crossroads. Another core WA-HMM Boxplot, which used balance efficiency accuracy...

The current light field occlusion removal methods are generally computationally demanding and have insufficient effect on the global receptive field. To address these issues, we introduce SwinSccNet, an network based Swin-Unet encoder–decoder system. We employ Scconv to compress redundant features in shallow convolutional neural (CNN), Swin transformer is used improve of deep encoder–decoder. experimental results show that our technique not only minimizes computational costs complexity but...

Video based object segmentation (VBOS) is an important step in many computer vision and multimedia tasks such as video editing compositing. In recent years, multi-view VBOS systems have become more popular because the stereo clues from multiview data can be efficiently incorporated to improve results eliminate required initial user input. this paper, we give a review on development of related techniques including acquisition, camera calibration, depth reconstruction, tracking. Furthermore,...

Abstract Lighting is crucial for portrait photography, yet the complex interactions between skin and incident light are expensive to model computationally in graphics difficult reconstruct analytically via computer vision. Alternatively, allow fast controllable reflectance lighting editing, we developed a physically based decomposition through deep learned priors from path-traced images. Previous approaches that used simplified material models or low-frequency low-dynamic-range struggled...

The limitations of task-specific and general image restoration methods for specific degradation have prompted the development all-in-one techniques. However, diversity patterns among multiple degradation, along with significant uncertainties in mapping between degraded images different severities their corresponding undistorted versions, pose challenges to tasks. To address these challenges, we propose Perceive-IR, an restorer designed achieve fine-grained quality control that enables...

Object segmentation plays an important role in multi-view video analysis. In this paper, we present a new object method for which only the disparity is used and motion estimation neglected. Firstly, modified locally adaptive support-weight approach proposed estimation. Then, realized by mean-shift algorithm. The experimental results show that could segment semantically meaningful objects from complex background with high precision.