120 results about "Photo image" patented technology

The invention relates to the technical field of image generation and image processing, in particular to a figure group photo generation method based on a depth generation model. The method aims at solving the problems that in the prior art, multi-person group photo generation depends on multiple pieces of template data, character features (such as hair styles and body types) are reduced insufficiently, the model collaboration efficiency is low, and the input image quality is sensitive. According to the main scheme, the method comprises the steps of 1, obtaining a target person photo to be subjected to group photo generation, and performing intelligent photo optimization processing on the target person photo to obtain an optimized person photo; 2, inputting the optimized character photo and a description cue word into an OmniGen model to generate a preliminary character group photo; and step 3, performing facial feature extraction and refined redrawing on the target person area in the preliminary person group photo by using a face replacement model to obtain a fidelity person group photo image with the similarity with the face of the target person being greater than 90%.

The invention discloses a method and system for constructing a fundus anomaly recognition model and a medium, and the method comprises the steps: collecting a normal fundus color image, constructing a training set and a verification set according to the normal fundus color image, and training an AutoEncoder model, thereby obtaining a basic pre-training model with the capability of reconstructing the normal fundus image; constructing a Unet model, freezing an encoder in the basic pre-training model, and fusing the Unet model with the frozen basic pre-training model to obtain a fused model; generating a mask matrix by using Berlin noise, and carrying out pixel replacement on the plurality of normal fundus color images based on the mask matrix to obtain a plurality of abnormal color images with abnormal areas; and performing data training on the fusion model by using the abnormal color photo image and the normal fundus color photo image to obtain a fundus anomaly recognition model. The fundus anomaly recognition model constructed by the invention can effectively segment and recognize the abnormal region in the fundus image, thereby greatly reducing the workload of doctors and improving the diagnosis efficiency and accuracy.

An electronic device, a method and a computer program product for controlling multiple cameras. The method includes detecting, via at least one processor, activation of a first camera of an electronic device to capture a first image incorporating a first region of interest within a field of view of the first camera and in response to detecting selection of a dual camera superimposed selfie capture mode for completing the image capture, capturing, via the first camera, the first image that includes the first region of interest. The method further includes contemporaneously capturing, via a second camera, a second image containing at least one face, cropping a facial image comprising the at least one face from the second image and superimposing the facial image onto the first image to generate a first composite image.

An information processing terminal according to an embodiment includes a determination unit, a display control unit, an acquisition unit, and a transmission unit. The determination unit determines whether information regarding a registration of a product satisfies a photographing condition. The display control unit allows a display unit to display an automatic photographing guidance for the product based on a determination result of the determination unit. The acquisition unit acquires, based on automatic photographing of the product, photographed image data of the photographed product. The transmission unit transmits the photographed image data acquired by the acquisition unit.

The invention provides an image synthesis method and device, equipment and a medium, and the method comprises the steps: carrying out the image embedding vector extraction of a target template image corresponding to a target face image and a target identification photo image based on an image embedding module in an image synthesis model, and obtaining an image embedding matrix; based on a feature processing module in the image synthesis model, performing semantic feature extraction, feature fusion and feature coding on the image embedding matrix in sequence to obtain a coding feature vector; and based on a diffusion model in the image synthesis model, performing image synthesis by applying a condition control vector formed by the coding feature vector to obtain the target identification photo image. According to the method, image generation is guided by image semantic features, it is effectively ensured that the synthesized target identification photo image not only accurately retains the rich-detail features of the face area of the target face image, but also meets the standard requirements in the target template image, and thus the quality and efficiency of image synthesis are improved.

An electronic punching bag device for enhancing training for a boxer includes a punching bag which includes a plurality of chains for suspending the punching bag from the suspension for a boxer to deliver punches to the punching bag. A display screen is wrapped around the punching bag to display imagery that is visible to the boxer. A communication unit is integrated into the punching bag and the communication unit receives data from a personal electronic device comprising photographic imagery or video imagery which is displayed on the display screen. A plurality of pressure sensors is each integrated into the punching bag to sense the pressure of the punches delivered by the boxer.

Systems and methods for generating a personalized three-dimensional avatar model are disclosed. A method includes generating pose data by identifying key points on a body figure mapped on user photographic images, loading a three-dimensional avatar model template based on persona characteristics and aligning the avatar model template with pose data to position the avatar in a corresponding posture. The personalized avatar model is generated using gradient descent optimization to adjust avatar model parameters based on a comparison of aligned avatar model with user images.

The invention discloses an image generation method and system based on artificial intelligence. The image generation method comprises the following steps: acquiring a content outline of a to-be-generated image by using basic information input by a user; acquiring information of a plurality of roles based on the content outline; obtaining makeup picture images of all roles by using the role information; and inputting the split information obtained by the content outline and the make-up photo image into a ComfyUI workflow to generate all split scene graphs of the story outline. According to the method and the device, each sub-mirror scene generated by the AI is ensured to be consistent in character image and artistic style, and the drawing production efficiency is improved.

A method for generating a digital ID comprising capturing one or more user images of user's face, transmitting the one or more user images or data derived therefrom to a verification server and determining if the one or more images represent a live person. Capturing an ID image of a photo identification for the user and transmitting the ID image to the verification server. Comparing at least one of the one or more user images or data derived therefrom to the image of the user on the photo identification to determine if user in the one or more user images is the same person as the image on the photo identification. Responsive to a match from the comparing and a determination that a live person was captured in the one or more user images, generating a user digital ID, and transmitting the user digital ID to the user.

An image processing apparatus according to this embodiment includes: a specifying unit configured to specify, when parameter values of illumination parameters at a plurality of points corresponding to pixels of a photographed image obtained by photographing an object at a predetermined sampling time are compared with each other, a plurality of peak points each of which is a peak with respect to surrounding parameter values; a generation unit configured to generate, based on respective peak values of the parameter values at the specified plurality of peak points, the parameter values at a plurality of points between the peak points, and thereby generate a distribution of the parameter values; and a correction unit configured to correct the photographed image obtained at the predetermined sampling time by correcting the parameter values of at least one pixel of the photographed image.

The invention discloses a fusion algorithm for cross-modal face recognition, which comprises data description, a data set comprises data from 142 subjects, a total of 2556 pairs of image samples, each pair of samples is composed of a black-and-white image collected by a cab and a color image of an identification photo, the number of the images of two modals is 50%, and in order to ensure the practical applicability of a test set, the fusion algorithm is provided with a fusion algorithm for cross-modal face recognition. 20% of the image pairs of each subject are randomly selected as a test set to ensure balanced distribution of different illumination conditions, shooting angles and expression changes, in addition, all the images are manually screened, samples which are seriously blurred, shielded or insufficient in illumination are removed, and the data quality is ensured; the modeling step comprises preprocessing, face area detection and face standardization, face vector representation, distance measurement, learning rate scheduler construction, loss function and model training.

A method and system to verify carpool occupancy compliance for access to High Occupancy Vehicle (HOV) lanes, High Occupancy or Toll (HOT) lanes, or other vehicle-occupancy contingent rewards. Software and hardware devices are used with radio-frequency transmitter modules to capture one or more photo images of vehicle occupants and to perform boxed headcounts of humans in any given photo frame. Biometric signature detection is used to confirm the boxed headcounts and a realness algorithm to further confirm the genuineness of any human image. Occupancy compliance can be communicated directly to an appropriate regulatory body.

The invention provides an art photo image enhancement processing method, and belongs to the field of image enhancement. According to the method, a dynamically generated multi-scale fusion coefficient and a cross-domain interaction mechanism are utilized, a hybrid domain attention (HDAB) module is constructed to extract frequency domain and spatial domain feature information of an art photo image, and fusion and enhancement of high-frequency, low-frequency and multi-scale features are realized; through a normalization and contrast stretching method, in combination with a region saliency enhancement strategy, an adaptive contrast enhancement (ACEM) module is constructed, contrast enhancement is performed on artistic photo image features, weight information is dynamically generated, target region features are highlighted, and the method improves the visual effect of the artistic photo and enhances the artistic expressive force and detail definition of the artistic photo.

The embodiment of the invention relates to an image generation method and device, equipment and a storage medium. The method comprises the steps of obtaining at least one personal image in response to a target interaction operation; the portrait image comprises one or more to-be-processed target objects; determining a reference group photo image, wherein the number of reference objects in the reference group photo image is the same as that of the target objects; based on the portrait image, fusing the target object to a reference object in the reference group photo image to generate an initial group photo image; based on a target diffusion model corresponding to the target object, adjusting at least one object feature of a fusion object corresponding to the target object in the initial group photo image, and generating a target group photo image; the target diffusion model is obtained by performing fine tuning training on the initial diffusion model based on an image sample of the target object. In this way, the flexibility and diversity of personalized image generation are improved, and the authenticity and image quality of the generated personalized image are improved.

The invention relates to an image generation method and device, electronic equipment and a medium, and the method comprises the steps: obtaining a plurality of groups of images, and each group of images comprises at least one first sample image containing an object to be subjected to group photo; obtaining group photo image description information for describing a group photo generation effect of the object to be subjected to group photo contained in each group of images; and based on the multiple groups of images and the group photo image description information, generating a group photo image which contains the object to be subjected to group photo contained in each group of images and is matched with the group photo image description information. The whole group photo image generation process does not need to use a shooting device, the group photo image matched with the group photo image description information can be created based on the images of the multiple groups of different objects to be subjected to group photo and the group photo image description information input by the user, the degree of freedom and playability are high, the generated group photo image is close to the real shooting effect, and the user experience is improved. Even if the user and the relatives are separated, the group photo image of the user and the relatives can be generated, and the user experience is improved.

The invention provides a device and a method for acquiring multi-point chromaticity coordinate values on the surface of a shot object, and mainly relates to the fields of color vision, colorimetry, color communication, image processing, computer control, deep learning, applied mathematics and the like. The problems that the multi-point chromaticity coordinate value of the surface of an existing shot object is difficult to measure, inaccurate in measurement and high in measurement cost are solved, and the multi-point chromaticity coordinate value of the surface of the shot object is calculated by obtaining a group photo image of the shot object and a color block group under the irradiation of different spectral distribution light sources.

The present application relates to an image generation method and apparatus, a device, and a storage medium. The image generation method comprises: in response to a target interaction operation, acquiring at least one portrait image, the portrait image comprising one or more target objects to be processed; determining a reference group photo image, the number of reference objects in the reference group photo image being the same as that of the target objects; on the basis of the portrait image, fusing the target objects to the reference objects in the reference group photo image to generate an initial group photo image; and on the basis of a target diffusion model corresponding to the target objects, adjusting at least one object feature of fused objects corresponding to the target objects in the initial group photo image, to generate a target group photo image, the target diffusion model being obtained by performing fine-tuning training on an initial diffusion model on the basis of image samples of the target objects.

An electronic device, a method and a computer program product provide techniques for automating self-image capturing, via a camera, when the device is in an always on camera mode. The method includes determining if a captured image stream contains a face of a user. In response to determining the image stream contains the face of the first user, the method includes determining whether an eye gaze direction of the user corresponds to a direction of the camera. In response to determining the user is looking towards the direction of the camera, the method includes determining if the electronic device is in an orientation that corresponds with the user taking a self-image. In response to determining the electronic device is in a selfie-taking orientation, while the user is looking towards the direction of the camera, the method includes autonomously capturing, via the camera, a self-photo image and storing the self-photo image.

In order to monitor operation situations of site resources at a construction site efficiently, a measuring system uses a surveying apparatus 100 including a camera and a position-determining function using laser light, and the measuring system includes a photographing means for continuously photographing construction machines 201 to 204 which are site resources for performing operations at a construction site by the camera, a recognizing means recognizing the site resources in photographed images obtained by the photographing, a tracking means for tracking the image of the site resources recognized in the multiple photographed images obtained by the continuous photographing, and a position-determining means collimating to the site resources which are objects for the tracking, and determining the positions of site resources by the position-determining function, in which the determining of positions is performed multiple times at intervals.

1. The name of the design product: children AI object recognition camera (C62). 2. The use of the design product: for taking photos, image recognition, encyclopedia, etc. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view.

A domain adaptation framework for generating a 3D avatar generative adversarial network (GAN) that is capable of generating an avatar based on a single photo image. A 3D avatar GAN is generated by training a target domain using an artistic data set. Each artistic data set includes a plurality of source images, each source image being associated with a style type, such as a cartoon, cartoon, and comic. In some embodiments, the domain adaptation framework begins from a source domain that has been trained according to 3D GAN and a target domain that has been trained with 2D GAN. The framework fine-tunes the 2D GAN by training it with an artistic data set. The obtained 3D head portrait GAN generates a 3D artistic head portrait; and the editing module is used for executing semantic and geometric editing.

Embodiments of the invention provide a group photo image generation method and apparatus, an electronic device and a storage medium. The method comprises the steps of obtaining user images corresponding to at least two invited users; according to the number of invited users and group photo control information, a target group photo template is determined, and the group photo control information represents a group photo posture and / or a group photo formation; a group photo image is generated based on the target group photo template and the user image, the group photo image comprises at least two group photo objects, the target group photo template is used for generating the image background of the group photo image, and the user image is at least used for generating the object face of the group photo object. A target group photo template is determined according to the number of invited users and group photo control information, and then an object body and an object face are constructed in combination with a user image and the target group photo template, so that a group photo image is generated. Multiple persons do not need to photograph on site, the facial expression flaws of the group photo object in the group photo image can be avoided, and the imaging quality of the group photo image is improved.

The invention provides a three-dimensional restoration method and device for an object, a storage medium and electronic equipment, and the method is applied to the technical field of computers, and comprises the steps: generating a color image of a target object based on an archaeological drawing of the target object and a photo image of other objects of the same type as the target object, the method comprises the steps of obtaining a color image of a target object, constructing an initial three-dimensional model of the target object based on the color image, obtaining a plurality of key view angle projection structure diagrams for the target object based on a norm division rule of the target object, and carrying out line texture adjustment processing on the initial three-dimensional model based on the key view angle projection structure diagrams and the color image, and obtaining a restored three-dimensional model of the target object. According to the method, the patterns of the target object at other unknown angles can be generated through the division rule expansion of the object, and line texture adjustment processing is performed on the three-dimensional model, so that the overall shape structure, the appearance style and the ornamentation characteristics of the ancient object are restored.

The invention provides an identification photo glasses blue light elimination method and device and terminal equipment, and is suitable for the technical field of image processing, and the method comprises the steps: carrying out the portrait region segmentation and color space conversion processing of identification photo image information to be subjected to blue light elimination, obtaining blue component image information of a skin area, blue component image information of a glasses area, certificate illumination component image information and red component image information of a certificate photo to obtain pixel information of a blue light area; and determining target blue component gray value adjustment information and target brightness component gray value adjustment information according to the blue light area pixel information, the glasses area blue component image information and the certificate illumination component image information, so as to perform color space inverse conversion processing in combination with the red component image information of the certificate photo. And generating blue light elimination identification photo image information. According to the invention, the accuracy and effectiveness of recognition and elimination processing of the blue light area of the glasses are improved, and the image quality of the identification photo generated after blue light elimination processing is ensured.

It has a heat exchanger (12), the heat exchanger (12) has a pipe (14) for carrying a cooling medium, the pipe (14) is connected to a plurality of cooling fins (10) of the heat exchanger (12), has a fan (8) for conveying a volumetric air flow, has a heating device (16) for defrosting the heat exchanger (12), has a photographic imaging device (20), and has a control unit (22), the control unit (22) is set to detect the frozen state of the heat exchanger (12) and determine a defrosting time based on the evaluation of a first photographic image (56), the control unit (22) is set to control a defrosting operation performed by the heating device (16) based on the evaluation of a second photographic image (68), and the evaluation of the first photographic image (56) is different from the evaluation of the second photographic image (68).

Domain adaptation frameworks for producing a 3D avatar generative adversarial network (GAN) capable of generating an avatar based on a single photographic image. The 3D avatar GAN is produced by training a target domain using an artistic dataset. Each artistic dataset includes a plurality of source images, each associated with a style type, such as caricature, cartoon, and comic. The domain adaptation framework in some implementations starts with a source domain that has been trained according to a 3D GAN and a target domain trained with a 2D GAN. The framework fine-tunes the 2D GAN by training it with the artistic datasets. The resulting 3D avatar GAN generates a 3D artistic avatar and an editing module for performing semantic and geometric edits.

The present invention relates to a method of photo localization in a space using video motion analysis and, more specifically, to a method for calculating the spatial position of a photographic image by using a video filmed by a video camera and a photographic image captured by a photographic camera during filming of the video. According to the present invention, the method of photo localization in a space using video motion analysis has an advantage of effectively finding a relative position of the photographic image in a three-dimensional shape, which can be identified by using video image frames, of a filming space.

To provide an inspection assistance system which can assist the inspection task of a pharmacist and a tablet packing device.SOLUTION: An inspection assistance system (1) includes an inspection display processing part (211) which is configured to display, per unit of packing, a capture image of a tablet which is captured before the tablet dispensed from one or both of a tablet cassette (41) and a manual distribution unit (42) based on prescription data is packed into a packing material in a tablet packing device (4), and is configured to display a result of an inspection process which is executed based on the prescription data and identification information of the tablet included in the capture image of the tablet. Thus, it is possible to assist a pharmacist's inspection task carried out with respect to the tablets packed by the tablet packing device (4).SELECTED DRAWING: Figure 1

Methods of inspecting fan blades of a gas turbine engine for abnormalities are disclosed. One method utilizes a neural network to determine whether a photographic image of at least one fan blade depicts an abnormality of the at least one fan blade. Another method uses a dimensional reduction technique on one or more first photographic images that depict at least one first fan blade to obtain basis vectors, utilizes the basis vectors to obtain a reconstructed version of a second photographic image that depicts at least one second fan blade, and determines whether the at least one second fan blade includes an abnormality based on a difference between the second photographic image and the reconstructed version of the second photographic image. Another method determines whether a fan blade includes an abnormality based on contrast differences between adjacent areas of an image.

The invention provides an intelligent photo system based on crowd classification, and relates to the technical field of image processing. A crowd category instruction input by a user is received; in response to the crowd category instruction, obtaining a photo template set associated with the crowd category instruction, the photo template set comprising a thumbnail composed of a plurality of photo templates; obtaining a target template selected by a user based on a plurality of thumbnails in the photo template set; collecting an original image of a user; and performing attitude analysis based on the original image and the target template, generating an optimal mapping relation, and generating an intermediate image according to the optimal mapping relation. The reality sense and the visual effect of the synthesized image are greatly improved, and the adaptability and the robustness of the system are enhanced, so that high-quality photo images can be stably output under different scenes and requirements.