Abstract: Embodiments are disclosed for user-guided variable-rate compression. A method of user-guided variable-rate compression includes receiving a request to compress an image, the request including the image, a corresponding importance data, and a target bitrate, providing the image, the corresponding importance data, and the target bitrate to a compression network, generating, by the compression network, a learned importance map and a representation of the image, and generating, by the compressing network, a compressed representation of the image based on the learned importance map and the representation of the image.

Abstract: A cross domain generalization system for industrial artificial intelligence (AI) applications is disclosed. A target encoder subsystem obtains target data from a target machine product and generates lower dimensional data for obtained target data using a target artificial intelligence (AI) model. The generated lower dimensional data are corresponding to a plurality of target embeddings data. The target encoder subsystem further applies the plurality of target embeddings data into a source classifier AI model. A source classifier subsystem predicts a quality of the target machine product by generating class labels for each of the plurality of target embeddings data based on a result of the classifier AI model. The goal of the present invention is to learn features or representations such that the correlation with a label space is similar both in source and target domains while being invariant of data distributions.

Abstract: Techniques for media stream patching using dynamic frame prediction are described. A frame loss detector of a decoder can detect a frame loss condition of an input media stream and cause a request for a substitute frame to be sent to a frame predictor. The frame predictor uses one or more machine learning models to infer a substitute frame, which is inserted into the media stream for further processing and distribution.