368 results about "Visual appearance" patented technology

In some embodiments, a computer system displays a caption for a media item at different depths depending on the depth of the portion of the media item over which the caption is displayed. In some embodiments, a computer system displays a user interface element that includes information associated with the media item at different locations relative to the media item depending on attention of the user. In some embodiments, a computer system displays a user interface element that includes information associated with the media item with different visual appearances depending on visual characteristics of the portion of the media item over which the user interface element is displayed.

Embodiments described herein provide a video generation framework built on a decoupled multimodal cross-attention module to simultaneously condition the generation on both an input image and a text input. The video generation may thus be conditioned on the visual appearance of a target object reflected in the input image. In this way, zero-shot video generation may be achieved with little fine-tuning efforts.

A digital agent is pre-trained to be a digital proxy for a user. Taking on the persona (e.g., personality, mannerisms, preferences, knowledge, and in some cases, a realistic visual appearance and voice of the human), the digital agent can effectively act on behalf of the human. During a virtual meeting, the pre-trained digital agent can listen to what the team has to say, ask clarifying questions, answer questions on the human's behalf, and raise points the human would want the team to consider. Since the digital agent visually resembles, sounds like, and acts like the human, the digital agent appears much like other remote participants, thereby improving the meeting experience of the other attendees and facilitating meeting productivity in the absence of a human team member.

An image inpainting system is described that receives an input image that includes a masked region. From the input image, the image inpainting system generates a synthesized image that depicts an object in the masked region by selecting a first code that represents a known factor characterizing a visual appearance of the object and a second code that represents an unknown factor characterizing the visual appearance of the object apart from the known factor in latent space. The input image, the first code, and the second code are provided as input to a generative adversarial network that is trained to generate the synthesized image using contrastive losses. Different synthesized images are generated from the same input image using different combinations of first and second codes, and the synthesized images are output for display.

An electronic device is provided. The electronic device includes a touchscreen, at least one processor, and memory storing instructions that, when executed by the one or more processors individually or collectively, cause the electronic device to, display, on the touchscreen, a first object indicating an adjustable range of a first setting value and a first indicator, on the first object, for displaying information about the first setting value at a first position for the first indicator, wherein the first indicator is movable with respect to the first object to adjust the first setting value, display, on the touchscreen, a second object indicating an adjustable range of a second setting value and a second indicator, on the second object, displaying information about the second setting value, wherein the second indicator is movable with respect to the second object to adjust the second setting value, while a touch input is maintained on the first indicator to change a position of the first indicator for the first object to change the first setting value, increase a size of the first indicator and display the information about the first setting value corresponding to a changed position of the first indicator, at a second position for the first indicator, wherein the second position is different from the first position to avoid being hidden by the touch input, and while the size of the first indicator is increased in response to the touch input being maintained on the first indicator, change a visual appearance of at least one of the second object or the second indicator.

The present invention provides an article visual inspection method and a rationale-generative estimation method using a large vision–language model.　The article visual inspection method comprises: a pretraining step of performing context-sensitive pretraining by inputting information containing images and sentences into a large vision–language model; an additional training step of performing additional training by inputting training information on a good-quality group and training information on a poor-quality group of a plurality of articles into the large vision–language model; and an inspection step of inputting a visual appearance image for inspection into the large vision–language model and causing the large vision–language model to output an inspection result. The article visual inspection method is characterized in that the training information on a good-quality group includes visual appearance images of good-quality articles and text information about good-quality articles, the training information on a poor-quality group includes visual appearance images of poor-quality articles and text information about poor-quality articles, and the information to be inspected includes an image to be inspected and text information requesting an inspection result. The estimation method comprises: a pretraining step of performing pretraining by inputting a plurality of datasets into a large vision–language model, each dataset containing a single image paired with a sentence containing a specified subject's preference information regarding the image; and a result outputting step of inputting an image of an object to be estimated into the large vision–language model and causing the large vision–language model to output an estimation result regarding individual preference and a rationale. The specified subject's preference information includes a judgment of whether or not the subject prefers the single image and text information indicating a rationale for the judgment.

The invention discloses a target identification method and device in an infrared image, equipment and a storage medium. The method for identifying the target in the infrared image comprises the steps of training a deep learning identification model fusing vision and track features based on a marked target data set; time sequence visual features of the target are extracted from the continuous multiple frames of infrared images; based on the position change of the multi-frame detection frame, reversely deducing the motion trail of the target in the physical space, and extracting the time sequence change characteristics of the trail; and fusing the time sequence visual features of the target and the track time sequence change features to form a joint feature vector, and identifying and classifying the joint features of the target through a deep learning identification model. According to the method, a target trajectory feature modeling mechanism is introduced, continuous multi-frame infrared image feature extraction is combined, and visual appearance features and physical space motion features are fused, so that the distinguishing capability of targets with similar appearances such as an unmanned aerial vehicle and a flying bird in a motion behavior dimension is effectively enhanced, and the accuracy of recognizing a complex target in an infrared image is improved.

A necklace fitting system that is operable to demonstrate a plurality of exemplary necklace lengths on a user wherein the present invention employs only a portion of the precious metal necklace. The present invention includes a base chain member wherein the base chain member includes a first end and a second end. The base chain member has integrally formed thereon a plurality of connection members. A first connection member is proximate the first end of the bae chain member and the remaining connection members are equally spaced between the first connection member and the second end of the base chain member. A plurality of sample members are provided wherein the sample members are six inches in length and are manufactured of the desired precious metal and style that a user is desiring to inspect visual appearance ensuing donning thereof. Exemplary lengths ranging from fourteen to thirty inches are provided.

The invention belongs to the field of computer vision, vision-language matching detection and multi-mode target tracking, and discloses a self-adaptive vision-language matching detection and tracking method. The method comprises the following steps: firstly, constructing a vision-language multi-modal tracking large model by utilizing modal alignment and representation advantages of a pre-trained multi-modal large model; a visual context perception prompt learning algorithm is provided, visual context features are extracted from current search features, prompt learning is adopted to integrate the visual context features into the text coding process, and dynamic perception of the text features on visual changes is achieved; besides, a language self-adaptive implicit updating mechanism is designed, a learnable vector is embedded in a text coding process, and dynamic visual context clues are combined, so that implicit updating of text features in a tracking process is realized, a static language can be dynamically matched with a visual appearance, and the visual-language detection tracking performance is improved.

An unattended approach can increase the reproducibility and safety of the treatment as the chance of over / under treating of a certain area is significantly decreased. On the other hand, unattended treatment of uneven or rugged areas can be challenging in terms of maintaining proper distance or contact with the treated tissue, mostly on areas which tend to differ from patient to patient (e.g. facial area). Delivering energy via a system of active elements embedded in a flexible pad adhesively attached to the skin offers a possible solution. The unattended approach may include delivering of multiple energies to enhance a visual appearance.

Techniques for reference image based material retrieval are described that support identification of procedural materials based on visual features of input images. A processing device, for instance, receives an input image that has a particular visual appearance. The processing device generates a histogram representation of the input image that represents a color prominence of the input image and generates a color distribution based on the color prominence. The processing device leverages a vision language model to filter candidate procedural materials by a semantic similarity to the input image. The processing device then identifies a procedural material that has a visual similarity to the particular visual appearance by comparing the color distribution for the input image to color distributions associated with the filtered candidate procedural materials. In this way, the techniques described herein support efficient retrieval of procedural materials based on color and on semantic features of the input image.

The invention discloses an automatic production workpiece visual appearance defect detection device which comprises two racks, a fixing plate is fixedly arranged at the upper end of the center position of the rack at the rear end, a second sliding groove is formed in the center position of the rack at the front end, and the second sliding groove is of a penetrating structure. A driving plate is arranged on one side of the interior of the second sliding groove, the driving plate is in sliding fit with the second sliding groove, second hydraulic telescopic devices are fixedly arranged on the sides, close to the exterior of the rack, of the driving plate and the fixing plate correspondingly, and the telescopic ends of the second hydraulic telescopic devices extend to the other sides of the driving plate and the fixing plate; a first mounting base is fixedly arranged outside the telescopic end of the second hydraulic telescopic device, a first hydraulic telescopic device is fixedly arranged at the upper end of the first mounting base, and the telescopic end of the first hydraulic telescopic device extends to the position below the first mounting base. And the problem that workpieces with defects are difficult to treat in time is solved.

The invention discloses an SIC pin tooth substrate coding visual appearance automatic detection system, and relates to the technical field of SIC pin tooth substrate detection, the SIC pin tooth substrate coding visual appearance automatic detection system comprises a mounting box body, the inner wall of the mounting box body is fixedly connected with a conveying mechanism, and the inner wall of the mounting box body is provided with a transfer mechanism at one end of the conveying mechanism. According to the invention, through the arrangement of multiple cleaning mechanisms, cleaning can be carried out before coding is carried out on the surface of the SIC pin tooth substrate, the situation that dust remains on the surface of the SIC pin tooth substrate and influences subsequent surface visual detection on the SIC pin tooth substrate is avoided, and through the arrangement of a turn-over mechanism, a quality pat carrying mechanism and a quality pat conveying mechanism, after large-plane detection is carried out on the SIC pin tooth substrate, the surface of the SIC pin tooth substrate can be cleaned, and the surface of the SIC pin tooth substrate can be subjected to large-plane detection on the surface of the SIC pin tooth substrate. According to the SIC pin tooth substrate detection device, one side of an SIC pin tooth substrate is turned over and then transferred to the position below a CCD quality shooting instrument, the other side of the SIC pin tooth substrate is visually detected, the quality shooting carrying mechanism can place the SIC pin tooth substrate on different conveying belts according to the detection result, the SIC pin tooth substrate is sorted according to good and bad products, and compared with conventional manual detection, the detection efficiency is higher.

An unattended approach can increase the reproducibility and safety of the treatment as the chance of over / under treating of a certain area is significantly decreased. On the other hand, unattended treatment of uneven or rugged areas can be challenging in terms of maintaining proper distance or contact with the treated tissue, mostly on areas which tend to differ from patient to patient (e.g. facial area). Delivering energy via a system of active elements embedded in a flexible pad adhesively attached to the skin offers a possible solution. The unattended approach may include delivering of multiple energies to enhance a visual appearance.

The invention relates to the technical field of digital protection and exchange of cultural heritage, in particular to a visual information design optimization method for a non-genetic-cultural genetic map. Through fusion of multi-modal image features and cross-scale stratified analysis, a visual optimization system of a non-genetic-culture genetic map is constructed. RGB, depth and infrared multi-mode images are synchronously collected, and macroscopic, mesoscopic and microscopic multi-scale feature extraction is combined, so that deep association of pattern visual representation and cultural semantics is realized; a pre-training semantic model adaptive to the field and cross-modal correlation degree calculation are adopted, cultural symbols and pattern visual features in a spoken text of an inheritor are effectively fused, and the problem of semantic superficial layer of a traditional method is solved; based on an iterative verification mechanism of dynamic weight adjustment, through quantitative matching of feature attributes and semantic parameters, the consistency of culture genes with different scale features is ensured.

The invention provides a simulation scene generation method and device, electronic equipment and a readable storage medium, and the method comprises the steps: carrying out the grid modeling of a target road element based on the real physical data of the target road element, and obtaining a grid-modeled target road element; performing Gaussian three-dimensional modeling on the grid modeled target road elements to obtain an initial Gaussian primitive set of the target simulation scene; and inputting the initial Gaussian primitive set and the target road video data into a trained simulation scene optimization model, optimizing visual attributes of each Gaussian primitive in the initial Gaussian primitive set, and obtaining a target simulation scene corresponding to the target road video data. Therefore, the geometric structure and the visual appearance of the object in the scene are decoupled, the geometric accuracy of the object is ensured through the grid model, and the visual attribute of the object is optimized through the trained simulation scene optimization model, so that the generation confidence and the generation efficiency of the simulation scene can be effectively improved.

Using a pre-trained and fixed Vision Transformer (ViT) model as an external semantic prior, a generator is trained given only a single structure / appearance image pair as input. Given two input images, a source structure image and a target appearance image, a new image is generated by the generator in which the structure of the source image is preserved, while the visual appearance of the target image is transferred in a semantically aware manner, so that objects in the structure image are “painted” with the visual appearance of semantically related objects in the appearance image. A self-supervised, pre-trained ViT model, such as a DINO-VIT model, is leveraged as an external semantic prior, allowing for training of the generator only on a single input image pair, without any additional information (e.g., segmentation / correspondences), and without adversarial training. The method may generate high quality results in high resolution (e.g., HD).

The invention relates to the technical field of resin, in particular to a preparation method of high-whiteness resin, which comprises the following steps: respectively preparing a mesoporous SiO2 (at) Eu < 3 + > whitening agent and a TiO2 photonic crystal whitening agent; the preparation method comprises the following steps: mixing two whitening agents according to a weight ratio of 1: 1-1: 4, carrying out surface modification, and carrying out melt blending with a resin matrix; an intelligent control system is adopted to regulate and control technological parameters in real time, and it is ensured that the whiteness is larger than or equal to 96 and the thermal decomposition temperature is larger than or equal to 380 DEG C. According to the preparation method, a mesoporous SiO2 (at) Eu < 3 + > whitening agent and a TiO2 photonic crystal whitening agent have a synergistic effect, and a silane coupling agent is adopted for surface modification, so that the whitening agents are more evenly and stably dispersed in a resin matrix, and the ultrahigh whiteness larger than or equal to 96 can be achieved; meanwhile, the yellowing risk is reduced, and the visual appearance quality of a resin product is greatly improved.

The present application relates to the technical field of resin, and relates to a preparation method of high-whiteness resin, respectively preparing mesoporous SiO2@Eu 3+ The two whitening agents are mixed in a weight ratio of 1:1 to 1:4, surface modification is carried out, and then the whitening agents are melt-blended with a resin matrix; an intelligent control system is used to real-time regulate process parameters, so that the whiteness is greater than or equal to 96, and the thermal decomposition temperature is greater than or equal to 380 DEG C; the present application realizes the synergistic effect of the mesoporous SiO2@Eu 3+ The whitening agent and the TiO2 photonic crystal whitening agent synergistically act, and surface modification is carried out by using a silane coupling agent, so that the whitening agent is uniformly and stably dispersed in the resin matrix, the whiteness of the resin product is greater than or equal to 96, the risk of yellowing is reduced, and the visual appearance quality of the resin product is greatly improved.

The system obtains multiple views associated with a first multiplicity of messages, where a view among the multiple views includes a criterion, a visual appearance, and an action. The criterion defines a message to include in the view, and the action is configured to trigger automatically when the message is included in the view. The system applies the view to the second multiplicity of messages by: selecting a subset of messages among the second multiplicity of messages according to the criterion; presenting the subset of messages to the user according to the visual appearance associated with the view; and performing the action on the subset of messages.

Apparatus, system, and method embodiments are disclosed for illuminating clothing, fashion accessories, and other items using diffusion and reflection of light from a light source. A fashion accessory can include a diffusion layer to provide a unique visual appearance from the light source based on the diffusion layer's manipulation of photon wavelength. The diffusion layer can reside under an exterior shell of the fashion accessory and diffuse light from the light source, providing a unique design aspect to the fashion accessory. The fashion accessory can further include a light refraction chamber to direct light in a visually appealing way, providing variations in quality and texture of the light that an observer sees. Additional layers of material over the exterior shell can provide additional variations in quality and texture of the observed light.

Provided is an inexpensive raised woodgrain finish decorative material that exhibits design properties similar to those of actual wood by the effect of reproducing visual appearance and feeling and has a short production timeline. The raised woodgrain finish decorative material has a substrate, a first raised section provided on one surface of the substrate, and a second raised section provided on the one surface of the substrate in a different location than the location of the first raised section, wherein the length of the first raised section is longer than the length of the second raised section in the lengthwise direction of the substrate, and the height of the first raised section is higher than the height of the second raised section in the thickness direction of the substrate.

Systems and associated methods for controlling operation of loading dock equipment are described herein. In some embodiments, the system and associated methods can be used to control operation of loading dock equipment (e.g., a vehicle restraint, a dock door, a dock leveler, etc.) according to a sequence of operations. The sequence of operations can include different sub-sequences based on loading dock conditions. The system can include a display screen that sequentially presents a series of control elements that enable operation of the loading dock equipment. Additionally, the visual appearance and / or sequence of presentation of the control elements indicate the proper sequence of selection to the user, thereby reducing user confusion and simplifying the operation of the loading dock equipment. Some functionality of the control panel can be enabled or disabled based on a current level of authorization.

A system, related operating methods, and computer readable storage media are disclosed here. The disclosed subject matter relates to methods of providing augmented reality features in connection with presentation of video content. A disclosed method involves: causing presentation of a video program at a user device associated with a viewing user; obtaining metadata associated with the video program; processing a live video feed that includes the viewing user; generating augmented reality overlay content for the live video feed, wherein visual appearance of at least some of the augmented reality overlay content is influenced by the obtained metadata; and causing presentation of an enhanced version of the live video feed at the user device associated with the viewing user, the enhanced version of the live video feed comprising the augmented reality overlay content.

The invention relates to a cross-modal video clip retrieval method based on a dual sequence diagram, which comprises the following steps of: giving an unedited video with any length and a text description, inputting the video and the text description into a cross-modal video clip retrieval model for processing, and finally obtaining a fusion activation diagram of the video and the text description, the video clip which is most matched with the text description is obtained by fusing the activation graph. The invention innovatively provides a semantic enhanced dual time sequence graph framework for text video clip retrieval. According to the method, visual content and semantic information are decoupled, and an enhanced visual appearance graph and an enhanced semantic graph are respectively constructed, so that the time sequence dependency relationship among video clips is more accurately captured. By means of the decoupling mode, the model can better distinguish fragments with similar visual content but different semantics.

A light-reactive gifting system includes, in one embodiment, an illumination device adapted to emit UV light, an ornamental attachment attached to or that forms at least a portion of the flat top of the illumination device, paper that includes a first light-reactive component that reacts to the UV light and is characterized by having a first visual appearance when not reflecting the UV light and a second visual appearance when reflecting the UV light, and an attachment mechanism that is useable to attach the illumination device to the paper thereby enabling the UV light to be cast in the direction of the paper and causing each of the first and second visual appearances based on activating or deactivating the UV LEDs.

An example operation may include one or more of ingesting data from a plurality of systems, dividing the data into a plurality of bins based on binning configuration settings, identifying a bin of data among the plurality of bins which contains an anomaly at a point in time based on thresholds for the plurality of bins, identifying a different bin of data among the plurality of bins which does not contain the anomaly at the point in time based on the thresholds for the plurality of bins, generating a heat map comprising a plurality of display elements corresponding to the plurality of bins including a display element corresponding to the bin of data with the anomaly with a different visual appearance than a display element corresponding to the different bin which does not contain the anomaly, and rendering the heat map via a GUI.

The present technology generates a simplified version of a dynamic object by capturing sprites from segments of the complex dynamic object while the complex dynamic object is rendered. When the complex dynamic object is requested in an environment in which it is not desirable to display the complex dynamic object, the sprite information can be used to provide a simplified version of the dynamic object. The simplified version of the dynamic object can have a similar visual appearance but can be easier to render. The spite information for the segments of the dynamic object can be recorded in texture data, and the sprite information can be looked up and selected in a GPU for rendering.

A method for decorating a timepiece component, particularly a skeleton type watch dial. The component includes at least two different visual appearances, each visual appearance being the unique combination of a surface finish and of a colour. The decoration has the following steps: i) providing a component blank with a first unfinished surface (10a); ii) creating on said first unfinished surface a first termination possessing a first surface finish; iii) applying to the first termination one or more layers of atoms so as to give the first termination a first colour; iv) creating a second termination having at least one breaking of a corner on the blank in order to form at least one bevel or at least one chamfer having a second unfinished surface; v) creating on said second unfinished surface a second termination conferring a second surface finish, and vi) using electroplating, PVD or ALD, to deposit a metallic layer, preferably a layer of rhodium or of gold, on the second termination so as to give the second termination a second colour.

A method of providing anonymized data for facilitating reidentification in a visual tracking system, the method comprising: detecting, in images obtained from a plurality of image sources, subareas which each contain a tracking target; computing, for each subarea, a feature vector which represents a visual appearance of the tracking target therein; for a first / second subgroup of the image sources, providing first / second reidentification data items by anonymizing each feature vector using a predefined one-way function modified by a first / second tracking-rights token, and disclosing the first / second reidentification data items annotated with locations of the respective subareas to a first / second tracking client, wherein the second tracking-rights token is distinct from the first tracking-rights token; and preventing access to the feature vectors.