53 results about "Visual artifact" patented technology

In various examples, a simulation platform generates a simulated driving environment by processing road map data to derive the location of wear-related visual artifacts for sections of a lane surface. Using map data, the simulation platform generates texture maps for aesthetic road rendering, which are used to apply textures to a 3D polygon topology mesh. The simulation platform generates visual artifacts representing the wear and tear of the road surface based on a calculation of one or more lane feature distances derived from the map data. The simulation platform calculates distances associated with reference line data derived from an image to render a texture of one or more lane features.The spacing is used to determine how the appearance of the texel is adjusted to include wear-related visual artifacts when displayed on a lane of the simulated driving environment.

Proximity-based surface texture generation for simulated environment systems and applications is disclosed. In various examples, a simulation platform generates a simulated driving environment by processing road map data to infer locations of wear-related visual artifacts for portions of a roadway surface. The simulation platform uses the map data to generate texture maps for aesthetic road rendering, which are used to apply texture to the 3D polygon topology mesh. The simulation platform generates visual artifacts representing use and wear of the roadway surface based on calculating one or more distances from roadway lane features derived from the map data. The simulation platform calculates a distance from the one or more roadway lane features associated with reference line data derived from the image to render the texture. The distances are used to determine how to adjust the appearance of texels to include wear-related visual artifacts when rendering wear-related virtual artifacts on the roadway of the simulated driving environment.

The invention discloses a unified unsupervised depth forgery detection method based on prototype guidance and double hyperspheres, and the method comprises the following steps: S1, extracting visual artifact features generated by depth forgery, and achieving the generation of pseudo labels through the clustering of a Gaussian mixture model; s2, performing comparative learning through a category prototype of momentum updating; and S3, realizing effective fusion of a feature space and a geometric decision by respectively constructing independent hyper-spheres for real and forged samples, and constructing a dual-depth support vector data description framework. The system has the beneficial effects that the system is composed of three core modules: a visual artifact feature-based pseudo label generator provides a reliable supervision signal, and the visual artifact feature-based pseudo label generator provides a visual artifact feature-based pseudo label description framework; a prototype guided contrast learning (PGCL) module enhances the discrimination capability through a prototype of momentum update, and a dual-depth support vector data description (Dual-DeepSVDD) module constructs a dual-hyperspherical decision boundary of true and false samples, thereby realizing effective integration of feature learning and geometric decision.

Compensating for burn-in of static icons on an OLED display involves periodically sample display content to generate an aging factor map indicating a relative aging level of pixels associated with one or more static icons compared to an average aging level of the display. The aging factor map is analyzed to detect areas of static icon burn-in and determine the relative aging level of affected pixels. Based on the relative aging level, a burn-in compensation strategy is selected, ranging from disabling compensation for weak aging to aggressively applying compensation for high aging. The selected strategy is applied to adjust the luminance of the affected pixels using a formula that incorporates a brightness adjustment factor and low-frequency and high-frequency compensation maps. Aspects also can include gradual transitions between compensation states to minimize visual artifacts and a user interface for configuring and monitoring the burn-in compensation process.

The invention discloses a lightweight diffusion virtual fitting algorithm for a high-resolution clothing human body image. The algorithm specifically comprises the following steps: step 1, acquiring a disclosed high-resolution virtual fitting data set; 2, preprocessing the clothing image and the reference figure image in the high-resolution virtual fitting data set obtained in the step 1 to obtain a figure image irrelevant to clothing; 3, constructing a lightweight diffusion virtual fitting model aiming at the high-resolution clothing human body image, and designing a required loss function; and step 4, training the lightweight diffusion virtual fitting model for the high-resolution clothing human body image by using the high-resolution clothing human body data set in the step 1. According to the method, the problems of detail loss and visual artifacts in a high-resolution scene in a virtual fitting method based on the generative adversarial network are solved, and the defects that a traditional diffusion model is huge in volume, high in computing resource demand and difficult to retain local features are overcome.

Using a fixed group of pictures (GOP) size in video encoding significantly hinders compression efficiency due to its inability to adapt to the dynamic nature of video content. While encoding leverages spatio-temporal redundancy within a GOP for compression, a predetermined size fails to capture the varying complexity of scenes. This leads to wasted bits in low-motion segments and insufficient reference frame variation for high-motion areas, resulting in visual artifacts and reduced compression efficiency. To address this limitation, a GOP size recommendation engine involving machine learning models can determine frame-level GOP size recommendations based on pre-encoder frame statistics. The frame-level GOP size recommendations are used to adapt the GOP size for encoding video frames.

The invention discloses a bar code scanning image enhancement method and system based on deep learning, and relates to the technical field of deep learning and image processing, and the method comprises the steps: collecting a bar code image, carrying out the preprocessing, and generating a standardized and grayscale image; constructing a difference value and a gradient response diagram based on the grayscale image, obtaining an attention map according to the standardized image, performing linear superposition in combination with the difference value and the gradient response diagram, obtaining a combined attention map, performing masking and shielding, and generating a mask and a shielding map; carrying out image interception on the mask image and the joint attention image based on the mask image and the calculation external rectangular frame, then executing channel splicing, obtaining an input tensor, inputting the input tensor into a conditional generative adversarial network, and outputting a repaired image; according to the invention, natural transition between a restoration result and an original image is realized by using a pixel-level fusion strategy based on Gaussian weight, and common visual artifacts such as edge fault and brightness jump can be effectively avoided.

This disclosure provides methods, devices, and systems for navigating medical instruments. The present implementations more specifically relate to photometric image enhancement techniques for endoscopy. In some aspects, a machine learning model may be trained to infer an enhanced image from a low-quality image captured by the camera of an endoscope. As used herein, the term “low-quality image” refers to any image containing visual artifacts, obstructions, and / or other deficiencies. By contrast, an “enhanced image” is a digitally modified representation of a low-quality image that removes and / or corrects at least some of the visual artifacts, obstructions, or other deficiencies in the low-quality image. A controller for a medical system may extract information from the enhanced image based on one or more image processing operations and generate a graphical user interface (GUI) for navigating the instrument within the anatomy based at least in part on the information extracted from the enhanced image data.

This invention discloses an image restoration method based on a hybrid structured sparse model, specifically including the following steps: initializing the restored image and setting the number of iterations; constructing a matrix of similar image patch groups; establishing a hybrid structured sparse model; using the hybrid structured sparse model to sparsely encode each similar image patch group; reconstructing each similar image patch group; and restoring the entire image based on all reconstructed similar image patch groups. The image restoration method of this invention better reconstructs details such as edges and textures, and effectively suppresses unwanted visual artifacts, further improving the image restoration effect.

The invention relates to a remote sensing image automatic registration method and device, equipment, a medium and a program product. The method comprises the steps of determining an effective overlapping region and a non-overlapping region of a reference image and a to-be-registered image, adaptively selecting a global registration process or a block registration process according to the size of the effective overlapping region, and obtaining an image pair subjected to corresponding registration processing; performing screening to obtain feature matching point pairs meeting a preset confidence threshold; according to the feature matching point pairs, respectively executing geometric transformation estimation and image correction of an adaptive matching corresponding registration process on the effective overlapping region and the non-overlapping region; and under the condition that the block registration process is executed, intelligent splicing is executed through a global optimal seam fusion algorithm based on dynamic programming, and a second registration result image with visual seamless is generated. By adopting the method, the feature matching robustness and precision in a complex scene can be improved, visual artifacts of image splicing after registration are effectively eliminated, and visual seamless fusion is realized.

A wearable heads-up display (WHUD) reduces the prevalence of visual artifacts by employing a projector with a combiner including a plurality of combiner elements such as dichroic prisms. The projector is configured with one or more features that 1) reduce the amount of stray light generated at the combiner 2) change the path of the stray light so that the stray light is unable to exit the projector and thus is unable to create visual artifacts, or any combination thereof. By reducing the stray light that is generated and by changing the path of the stray light as described herein, the likelihood that a user will see a visual artifact is reduced, thus improving the user experience with the WHUD system.

This invention discloses a general robustness enhancement method for a circuit board solder joint defect identification model. First, training samples are input into a pre-trained general diffusion generation model to extract latent variables. A flexible penalty term based on ReLU properties is used to replace hard truncation to constrain the perturbation range, leveraging the strong structural prior of the diffusion model to ensure that the generated latent space adversarial samples highly conform to the distribution of real solder joint images. Second, a smoothing period is introduced during the adversarial optimization iteration of latent variables, and a strategy based on historical averages is used to periodically update the optimization trajectory to filter out high-frequency components, thereby avoiding visual artifacts caused by drastic fluctuations in latent variables and interference with the learning of real features. Finally, before calculating the adversarial loss, the generated samples are subjected to multi-scale scaling and multi-directional translation, and the target model weights are randomly zeroed multiple times to simulate the training blind zone, so that the calculation of the adversarial gradient covers the uncertainty of spatial geometric transformation and model dynamic evolution.

The invention provides an interpretable text image tampering detection method and system based on visual artifacts and semantic anomaly fusion. The method comprises the steps that a tampered text image and a tampering analysis report corresponding to the tampered text image are obtained, and data preprocessing is conducted on the tampered text image; constructing a tampering locator, and training the tampering locator based on the tampering text image and the tampering analysis report after data preprocessing; constructing a tampering interpreter, and training the tampering interpreter based on the tampering text image after data preprocessing and the tampering analysis report; and obtaining a tampering detection model based on the trained tampering locator and tampering interpreter, and inputting the to-be-detected image into the tampering detection model to realize tampering detection. According to the method, the visual expert model and the multi-modal large language model are used for cooperative work, accurate positioning and reason description of the tampered area are achieved, high detection precision and robustness are kept under degradation and cross-scene conditions such as compression and fuzziness, and high practicability is achieved.

A light emitting display can be formed from tiles mounted within a certain distance range with respect to each other and with an established blending region positioned towards the edges of the tiles. A tile can be a matrix of light emitting elements, such as LEDs, OLEDs, quantum dots, or other element that emits light. The tolerance of spacing between tiles can allow for less precision in alignment during installation in a theatre, thereby reducing display assembly cost but still maintaining a display for displaying an image at a high quality with reduced or eliminated appearance of visual artifacts between tiles.

A method according to an example embodiment includes determining at least one organ contour of medical imaging data, the medical imaging data comprise at least one image of at least a part of at least one organ; at least one of, determining an image uncertainty information describing at least one area of the at least one image with a visual artifact, determining an organ contour inaccuracy information describing if the determination of the at least one organ contour is inaccurate, or determining a warning information describing at least one of a predetermined warning concerning the organ or an area surrounding the organ for which the organ contour was determined; and determining review data comprising the at least one image and the determined at least one organ contour and at least one of the determined image uncertainty information, the determined organ contour inaccuracy information or the determined warning information.

An endoscopic system, endoscope, and method for its operation are disclosed. The system comprises a light source, an endoscope with a distal light entrance surface, an image sensor capturing images through the light entrance surface and a rinsing device for providing a rinsing fluid flow for rinsing the light entrance surface, the method including the steps of receiving a sensor signal from a sensor detecting a user input starting a rinsing operation of the rinsing device, and, upon receipt of the sensor signal, changing an illumination mode of the light source from a first illumination mode to a second illumination mode and / or changing the video capture from a first video capture mode to a second video capture mode. The second illumination and / or video capture mask undesirable visual artifacts which can occur when collecting images with a rolling shutter sensors while performing a liquid flush of the light entrance surface.

The invention discloses an image super-resolution reconstruction method based on a multi-scale mixed attention residual network, and belongs to the technical field of computer vision and image processing. The invention provides an end-to-end reconstruction network (MHARN) aiming at the problems of insufficient detail recovery capability, limited cross-dimension feature interaction and poor adaptability to different image contents in the prior art. The method comprises the following steps: firstly, dynamically aggregating multi-scale features through parallel convolution branches with different voidage by utilizing a progressive convolution group (PCG), and breaking through the limitation of a single receptive field; the features are input into a cascaded enhanced residual attention block (ERAB), a multi-head hybrid attention module (MHAM) in the ERAB is utilized to capture window texture, geometric structure and global channel information in parallel, and a dynamic feature enhancement module (DFEM) is combined to adaptively generate convolution kernel parameters according to local image content; and finally, a high-resolution image is obtained through global feature fusion and up-sampling reconstruction. The texture detail recovery capability can be effectively enhanced, the visual artifacts are remarkably reduced, and the image reconstruction quality is improved.

A diffractive waveguide device includes an optical waveguide and a diffractive element optically coupled to the optical waveguide. The diffractive element is configured to alter a polarization and propagation direction of light of a first polarization, and is configured to transmit light of a second polarization without substantially altering a polarization or propagation direction thereof. A polarizing film assembly is configured to provide the light of the second polarization to the optical waveguide, and / or is configured to block the light of the second polarization from the optical waveguide. The polarizing film assembly includes a polarizer and an optical retarder that is positioned between the polarizer and the optical waveguide. Related devices and methods of operation are also discussed.

Generating the 3D representation of the object includes obtaining sensor data of the object. Media assets including objects may be obtained from a digital asset repository. Visual artifacts of an object may be generated from a media asset and used with sensor data to generate one or more virtual representations of the object. The visual artifacts include texture and / or geometric characteristics of the visual appearance of the object and are derived from image data in the media asset.

In various examples, a simulation platform generates a simulated driving environment by processing road map data to infer the location of wear-related visual artifacts for portions of a roadway surface. Using map data, the simulation platform generates texture maps for aesthetic road renderings that are used to apply textures onto a 3D polygon topology mesh. The simulation platform generates visual artifacts representing use and wear of the roadway surface based on calculating one or more distances from roadway lane features derived from the map data. The simulation platform computes distances associated with reference line data derived from an image to render texture from one or more roadway lane features. The distances are used in determining how the appearance of the texels are adjusted to include the wear-related visual artifacts when rendered on a roadway of the simulated driving environment.

Techniques are described for an encoder or decoder to adaptively change (e.g., 604) code processing based on a particular user's sensitivity to flicker or flash or blocking or other visual artifacts. Alternatively, the video may be pre-processed (e.g., 1000) to suppress artifacts based on the sensitivity of the user prior to encoding, and / or post-processed to suppress artifacts after decoding.

This invention discloses a method where volumetric content is encoded into a cluster by an encoder and transmitted to a decoder that retrieves the volumetric content. Clusters shared across different viewpoints are obtained and collectively managed. These clusters are then projected onto a 2D image and encoded into independent video streams. This achieves reduction in visual artifacts, as well as reduction in storage and streaming data.

A method includes receiving content for presentation on a display, and obtaining a plurality of image tiles from the content. Each image tile is segmented from a bottom horizontal center portion of a subsampled grayscale image of a frame of content. The method further includes detecting a region of interest of the moving text in the content by applying a search algorithm to the plurality of image tiles. A search algorithm utilizes the first and second sets of features to detect the region of interest with respect to the vertical and horizontal axes of the display, respectively. The method also includes correcting one or more visual artifacts in the region of interest, wherein the one or more visual artifacts include corrupted text.

Certain aspects of the present disclosure provide techniques for rendering visual artifacts in virtual worlds using machine learning models. An example method generally includes identifying, based on a machine learning model and a streaming natural language input, an intent associated with the streaming natural language input; generating, based on the identified intent associated with the streaming natural language input, one or more virtual objects for rendering in a virtual environment displayed on one or more displays of an electronic device; and rendering the generated one or more virtual objects in the virtual environment.

The application discloses a speech video generation method based on audio and video structure alignment and belongs to the virtual digital person field.The application comprises an audio segmentation module, an audio conversion module, an audio coding module, a video coding module and a video fusion decoding module.The audio conversion module is used for converting segmented phonemes into a mel-frequency spectrogram which is more in line with the frequency range of human ears according to Fourier transform.In the audio coding process, the frames of the same phonemes are taken as a continuous time module, and the time module is taken as time consistency to constrain the change of the lips, so that the fine-grained control of the lips is realized at the phoneme level through the time consistency constraint.In the video coding, the part region of the multi-pose change face in the input video is set as a mask region, the mask region is used as spatial consistency to accurately control the change amplitude of the lips, the position of the speaker's lips is aligned, the visual artifacts of the video are reduced, the facial details are optimized, and a high-quality speech video with audio-visual synchronization is generated.

A neural-network is trained to produce a visibility model based on just noticeable difference training data. The trained network is used to generate visibility masks to control the visual quality of image processing operations, such as digital watermarking, compression and other image editing operations. The visibility model is refined by iteratively adjusting for over and under predictions of a just noticeable difference threshold via an interactive user interface that pinpoints visual artifacts and captures updates from users, who provide adjustments such that the artifact is below what they perceive to be a just noticeable difference. Image processing, such as digital watermarking is performed in real time to provide the user with feedback to ascertain just noticeable difference levels corresponding to watermark embedding operations.

This application relates to a method, apparatus, device, medium, and program product for automatic registration of remote sensing images. The method includes: determining the effective overlapping and non-overlapping regions of a reference image and an image to be registered; adaptively selecting a global registration process or a block registration process based on the size of the effective overlapping region to obtain corresponding image pairs for registration processing; filtering feature matching point pairs that meet a preset confidence threshold; performing geometric transformation estimation and image correction adapted to the corresponding registration process on the effective overlapping and non-overlapping regions based on the feature matching point pairs; and, in the case of executing the block registration process, performing intelligent stitching through a globally optimal seam fusion algorithm based on dynamic programming to generate a visually seamless second registration result image. This method can improve the robustness and accuracy of feature matching in complex scenes, effectively eliminate visual artifacts in the stitching of registered images, and achieve visually seamless fusion.

The invention discloses a living cell grid protein imaging parameter optimization method based on an SRRF nanometer microscope and a related device, and relates to the field of biomedical imaging, the method comprises the following steps: under a living cell culture condition, using a TIRF microscope to carry out time sequence imaging on a living cell sample expressing a fluorescence label, and obtaining an original image sequence; inputting the plurality of core reconstruction parameters into a NanoJ-SRRF processing system, systematically combining the plurality of core reconstruction parameters, and generating a plurality of super-resolution reconstruction images; for the super-resolution image reconstructed by each combination parameter, a multi-modal evaluation system is adopted for comprehensive quantitative evaluation, and a multi-modal quantitative quality evaluation result is obtained; and based on the result, screening out a parameter combination which meets the preset resolution and the structure fidelity index and has the minimum visual artifact, and determining the parameter combination as an optimal parameter set of living cell grid protein coated small nest imaging. According to the invention, reconstruction artifacts can be reduced, and high-fidelity nanoscale observation of living cell CCPs dynamics is realized.

The invention relates to a makeup area picture consistency processing method and system based on a multi-mirror image, and the method comprises the steps: obtaining an auxiliary picture shot by a camera in a movable makeup mirror, and carrying out the face feature recognition and makeup behavior recognition of the auxiliary picture; after image parameters of the important makeup areas are extracted, the image parameters are applied to correction of the corresponding makeup areas in the main picture of the fixed makeup mirror, so that the video picture effects of the important makeup areas of the main picture and the auxiliary picture are consistent; further performing smoothing processing on the corrected makeup area and the peripheral area thereof, eliminating visual artifacts caused by sudden change of image parameters, and realizing smooth transition of boundaries; according to the scheme, personalized adjustment can be carried out according to the actual make-up behavior and the key area of the user, the consistency of make-up video pictures at different angles in the intelligent make-up mirror is effectively improved, and thus the accuracy of make-up video shooting is improved.

The invention provides an image processing method and device and a storage medium, and the method comprises the steps: carrying out the depth estimation of an initial image, and obtaining an initial depth map; performing expansion processing based on the initial depth map to obtain a target depth map texture; and rendering based on the viewpoint offset determined by the interaction event, the target depth map texture and the initial image to obtain a target rendering frame. According to the method, the initial depth map obtained by estimating the depth of the initial image is subjected to expansion processing, so that one or more pixels are expanded from a region with brighter pixels to a region with darker pixels, and a depth region slightly larger than the visual contour of an object is formed; and then, rendering is performed by using the viewpoint offset, the target depth map texture and the initial image to determine a target rendering frame, so that the problem of visual artifacts during rendering can be avoided.