Abstract: Systems and methods for training a visual question answer model include training a teacher model by performing image conditional visual question generation on a visual language model (VLM) and a targeted visual question answer dataset using images to generate question and answer pairs. Unlabeled images are pseudolabeled using the teacher model to decode synthetic question and answer pairs for the unlabeled images. The synthetic question and answer pairs for the unlabeled images are merged with real data from the targeted visual question answer dataset to generate a self-augmented training set. A student model is trained using the VLM and the self-augmented training set to return visual answers to text queries.

Abstract: A method for training a model for face recognition is provided. The method forward trains a training batch of samples to form a face recognition model w(t), and calculates sample weights for the batch. The method obtains a training batch gradient with respect to model weights thereof and updates, using the gradient, the model w(t) to a face recognition model what(t). The method forwards a validation batch of samples to the face recognition model what(t). The method obtains a validation batch gradient, and updates, using the validation batch gradient and what(t), a sample-level importance weight of samples in the training batch to obtain an updated sample-level importance weight. The method obtains a training batch upgraded gradient based on the updated sample-level importance weight of the training batch samples, and updates, using the upgraded gradient, the model w(t) to a trained model w(t+1) corresponding to a next iteration.

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: 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: Methods and systems for training a trajectory prediction model and performing a vehicle maneuver include encoding a set of training data to generate encoded training vectors, where the training data includes trajectory information for agents over time. Trajectory scenarios are simulated based on the encoded training vectors, with each simulated trajectory scenario representing one or more agents with respective agent trajectories, to generate simulated training data. A predictive neural network model is trained using the simulated training data to generate predicted trajectory scenarios based on a detected scene.

Abstract: Methods and systems for detecting faults include capturing an image of a scene using a camera. The image is embedded using a segmentation model that includes an image branch having an image embedding layer that embeds images into a joint latent space and a text branch having a text embedding layer that embeds text into the joint latent space. Semantic information is generated for a region of the image corresponding to a predetermined static object using the embedded image. A fault of the camera is identified based on a discrepancy between the semantic information and semantic information of the predetermined static image. The fault of the camera is corrected.

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: Systems and methods for generating a hypernetwork configured to be trained for a plurality of tasks; receiving a task preference vector identifying a hierarchical priority for the plurality of tasks, and a resource constraint as a tuple; finding tree sub-structures and the corresponding modulation of features for every tuple within an N-stream anchor network; optimizing a branching regularized loss function to train an edge hypernet; and training a weight hypernet, keeping the anchor net and the edge hypernet fixed.

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 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.

Abstract: A computer-implemented method for rut detection is provided. The method includes detecting, by a rut detection system, areas in a road-scene image that include ruts with pixel-wise probability values, wherein a higher value indicates a better chance of being a rut. The method further includes performing at least one of rut repair and vehicle rut avoidance responsive to the pixel-wise probability values. The detecting step includes performing neural network-based, pixel-wise semantic segmentation with context information on the road-scene image to distinguish rut pixels from non-rut pixels on a road depicted in the road-scene image.

Abstract: A method for jointly removing rolling shutter (RS) distortions and blur artifacts in a single input RS and blurred image is presented. The method includes generating a plurality of RS blurred images from a camera, synthesizing RS blurred images from a set of GS sharp images, corresponding GS sharp depth maps, and synthesized RS camera motions by employing a structure-and-motion-aware RS distortion and blur rendering module to generate training data to train a single-view joint RS correction and deblurring convolutional neural network (CNN), and predicting an RS rectified and deblurred image from the single input RS and blurred image by employing the single-view joint RS correction and deblurring CNN.

Abstract: A computer-implemented method for implementing face recognition includes obtaining a face recognition model trained on labeled face data, separating, using a mixture of probability distributions, a plurality of unlabeled faces corresponding to unlabeled face data into a set of one or more overlapping unlabeled faces that include overlapping identities to those in the labeled face data and a set of one or more disjoint unlabeled faces that include disjoint identities to those in the labeled face data, clustering the one or more disjoint unlabeled faces using a graph convolutional network to generate one or more cluster assignments, generating a clustering uncertainty associated with the one or more cluster assignments, and retraining the face recognition model on the labeled face data and the unlabeled face data to improve face recognition performance by incorporating the clustering uncertainty.

Abstract: Systems and methods for obstacle 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 one or more road scenes having obstacles. 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 for construction zone segmentation 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 construction zones scenes having various objects. 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: A computer-implemented method for implementing face recognition includes receiving training data including a plurality of augmented images each corresponding to a respective one of a plurality of input images augmented by one of a plurality of variations, splitting a feature embedding generated from the training data into a plurality of sub-embeddings each associated with one of the plurality of variations, associating each of the plurality of sub-embeddings with respective ones of a plurality of confidence values, and applying a plurality of losses including a confidence-aware identification loss and a variation-decorrelation loss to the plurality of sub-embeddings and the plurality of confidence values to improve face recognition performance by learning the plurality of sub-embeddings.

Abstract: A method for protecting visual private data by preventing data reconstruction from latent representations of deep networks is presented. The method includes obtaining latent features from an input image and learning, via an adversarial reconstruction learning framework, privacy-preserving feature representations to maintain utility performance and prevent the data reconstruction by simulating a black-box model inversion attack by training a decoder to reconstruct the input image from the latent features and training an encoder to maximize a reconstruction error to prevent the decoder from inverting the latent features while minimizing the task loss.