Abstract: A method for improving face recognition from unseen domains by learning semantically meaningful representations is presented. The method includes obtaining face images with associated identities from a plurality of datasets, randomly selecting two datasets of the plurality of datasets to train a model, sampling batch face images and their corresponding labels, sampling triplet samples including one anchor face image, a sample face image from a same identity, and a sample face image from a different identity than that of the one anchor face image, performing a forward pass by using the samples of the selected two datasets, finding representations of the face images by using a backbone convolutional neural network (CNN), generating covariances from the representations of the face images and the backbone CNN, the covariances made in different spaces by using positive pairs and negative pairs, and employing the covariances to compute a cross-domain similarity loss function.

Abstract: A display driving integrated circuit (IC) and a driving parameter adjustment method thereof are provided. The display driving IC includes a control circuit and a driving parameter determination circuit. The control circuit controls a current driving circuit and a scanning circuit according to a driving parameter, wherein the current driving circuit is suitable for driving multiple driving lines of a light emitting diode (LED) array, and the scanning circuit is suitable for driving multiple scanning lines of the LED array. The driving parameter determination circuit is coupled to the control circuit to provide the driving parameter. The driving parameter determination circuit dynamically adjusts the driving parameter for a target LED in the LED array according to a grayscale value of the target LED.

Abstract: Provided is a method for diagnosing and monitoring progression of cancer or effectiveness of a therapeutic treatment. The method includes detecting a methylation level of at least one gene in a biological sample containing circulating free DNA. Also provided are primer pairs and probes for diagnosis or prognosis of cancer in a subject in need thereof.

Abstract: A method of manufacturing a semiconductor device is provided. The method includes placing a package substrate on a carrier substrate, forming a frame on the package substrate, and affixing an active side of a semiconductor die on the package substrate. The semiconductor die together with the frame and the package substrate form a cavity between the semiconductor die and the package substrate. At least a portion of the semiconductor die and the package substrate are encapsulated with an encapsulant. The frame is configured to prevent the encapsulant from entering the cavity.

Abstract: Video methods and systems include extracting features of a first modality and a second modality from a labeled first training dataset in a first domain and an unlabeled second training dataset in a second domain. A video analysis model is trained using contrastive learning on the extracted features, including optimization of a loss function that includes a cross-domain regularization part and a cross-modality regularization part.

Abstract: Video methods and systems include extracting features of a first modality and a second modality from a labeled first training dataset in a first domain and an unlabeled second training dataset in a second domain. A video analysis model is trained using contrastive learning on the extracted features, including optimization of a loss function that includes a cross-domain regularization part and a cross-modality regularization part.

Abstract: Video methods and systems include extracting features of a first modality and a second modality from a labeled first training dataset in a first domain and an unlabeled second training dataset in a second domain. A video analysis model is trained using contrastive learning on the extracted features, including optimization of a loss function that includes a cross-domain regularization part and a cross-modality regularization part.

Abstract: A face model building method is provided. The face model building method includes: obtaining a plurality of facial feature animation objects and a plurality of object parameters respectively corresponding to the facial feature animation objects, where the facial feature animation objects include a plurality of two-dimensional facial feature animation objects and at least one three-dimensional facial feature animation object; and integrating the facial feature animation objects according to the object parameters to generate a three-dimensional face model. A face model building system is further provided.

Abstract: A face model editing method adapted to a face model editing system having a modeling platform and an editing platform is provided. The modeling platform has a plurality of face feature animation objects and a plurality of object parameters thereof. The face model editing method includes: receiving an object selection instruction by using the editing platform, and accessing the object parameter of the face feature animation object from the modeling platform according to the object selection instruction; receiving an adjusting instruction by using the editing platform, and adjusting the accessed object parameter; transmitting, by the editing platform, the adjusted object parameter to the modeling platform to update the object parameters; and generating, by the modeling platform, a three-dimensional face model by using the updated object parameters in combination with the face feature animation objects, and transmitting the three-dimensional face model to the editing platform for demonstration.

Abstract: An augmented reality implementing method applied to a server, which includes a plurality of augmented reality objects and a plurality of setting records corresponding to the augmented reality objects respectively is provided. Firstly, the server receives an augmented reality request from a mobile device, where the augmented reality request is related to a target device. Then, the server is communicated with the target device to access current information. Then, the server determines the current information corresponds to which one of the setting records, and selects one of the augmented reality objects based on the determined setting record as a virtual object provided to the mobile device.

Abstract: Methods and systems for training a neural network include generate an image of a mask. A copy of an image is generated from an original set of training data. The copy is altered to add the image of a mask to a face detected within the copy. An augmented set of training data is generated that includes the original set of training data and the altered copy. A neural network model is trained to recognize masked faces using the augmented set of training data.

Abstract: A scanner is provided and includes a load bearing element, a scan assembly, a transparent plate and a casing. The load bearing element includes a support element and a base. The support element is disposed on the base and surrounds the base. The scan assembly is configured to perform a scan operation and is disposed on the base. The transparent plate is disposed on the support element and is configured to bear an object. The casing includes a first casing part and a second casing part. The second casing part includes a bottom plate and at least one cylinder. The at least one cylinder is disposed on the bottom plate. The load bearing element is disposed on the at least one cylinder, and is located between the first casing part and the second casing part. A first buffer space is formed between the base and the bottom plate.

Abstract: A scanner is provided and includes a load bearing element, a scan assembly, a transparent plate and a casing. The load bearing element includes a support element and a base. The support element is disposed on the base and surrounds the base. The scan assembly is configured to perform a scan operation and is disposed on the base. The transparent plate is disposed on the support element and is configured to bear an object. The casing includes a first casing part and a second casing part. The second casing part includes a bottom plate and at least one cylinder. The at least one cylinder is disposed on the bottom plate. The load bearing element is disposed on the at least one cylinder, and is located between the first casing part and the second casing part. A first buffer space is formed between the base and the bottom plate.

Abstract: A method for achieving high-fidelity novel view synthesis and 3D reconstruction for large-scale scenes is presented. The method includes obtaining images from a video stream received from a plurality of video image capturing devices, grouping the images into different image clusters representing a large-scale 3D scene, training a neural radiance field (NeRF) and an uncertainty multilayer perceptron (MLP) for each of the image clusters to generate a plurality of NeRFs and a plurality of uncertainty MLPs for the large-scale 3D scene, applying a rendering loss and an entropy loss to the plurality of NeRFs, performing uncertainty-based fusion to the plurality of NeRFs to define a fused NeRF, and jointly fine-tuning the plurality of NeRFs and the plurality of uncertainty MLPs, and during inference, applying the fused NeRF for novel view synthesis of the large-scale 3D scene.

Abstract: A computer-implemented method for model training is provided. The method includes receiving, by a hardware processor, sets of images, each set corresponding to a respective task. The method further includes training, by the hardware processor, a task-based neural network classifier having a center and a covariance matrix for each of a plurality of classes in a last layer of the task-based neural network classifier and a plurality of convolutional layers preceding the last layer, by using a similarity between an image feature of a last convolutional layer from among the plurality of convolutional layers and the center and the covariance matrix for a given one of the plurality of classes, the similarity minimizing an impact of a data model forgetting problem.

Abstract: Three-dimensional occlusion can be used when generating AR display overlays. Depth information can be used to delete portions of an AR element, based on intervening objects between a viewer and the AR element. In cases where the depth information does not impart a complete picture of the intervening objects, additional image processing and object detection systems and techniques can be used to further improve the precision of the occlusion.

Abstract: Systems and methods for human detection are provided. The system aligns image level features between a source domain and a target domain based on an adversarial learning process while training a domain discriminator. The target domain includes humans in one or more different scenes. The system selects, using the domain discriminator, unlabeled samples from the target domain that are far away from existing annotated samples from the target domain. The system selects, based on a prediction score of each of the unlabeled samples, samples with lower prediction scores. The system annotates the samples with the lower prediction scores.

Abstract: Systems and methods are provided for multi-modal test-time adaptation. The method includes inputting a digital image into a pre-trained Camera Intra-modal Pseudo-label Generator, and inputting a point cloud set into a pre-trained Lidar Intra-modal Pseudo-label Generator. The method further includes applying a fast 2-dimension (2D) model, and a slow 2D model, to the inputted digital image to apply pseudo-labels, and applying a fast 3-dimension (3D) model, and a slow 3D model, to the inputted point cloud set to apply pseudo-labels. The method further includes fusing pseudo-label predictions from the fast models and the slow models through an Inter-modal Pseudo-label Refinement module to obtain robust pseudo labels, and measuring a prediction consistency for the pseudo-labels.

Abstract: Systems and methods for domain adaptation for structured output via disentangled representations are provided. The system receives a ground truth of a source domain. The ground truth is used in a task loss function for a first convolutional neural network that predicts at least one output based on inputs from the source domain and a target domain. The system clusters the ground truth of the source domain into a predetermined number of clusters, and predicts, via a second convolutional neural network, a structure of label patches. The structure includes an assignment of each of the at least one output of the first convolutional neural network to the predetermined number of clusters. A cluster loss is computed for the predicted structure of label patches, and an adversarial loss function is applied to the predicted structure of label patches to align the source domain and the target domain on a structural level.

Abstract: Systems and methods for lane marking and road sign recognition are provided. The system aligns image level features between a source domain and a target domain based on an adversarial learning process while training a domain discriminator. The target domain includes one or more road scenes having lane markings and road signs. The system selects, using the domain discriminator, unlabeled samples from the target domain that are far away from existing annotated samples from the target domain. The system selects, based on a prediction score of each of the unlabeled samples, samples with lower prediction scores. The system annotates the samples with the lower prediction scores.