The present invention relate generally to signal encoding and decoding. One claim recites a method comprising: obtaining color image data or color video data, the color image data or color video data comprising an encoded signal pattern, the encoded signal pattern aiding detection of an encoded message, the pattern comprising first frequency components and second frequency components, the color image data or color video data comprising first color data and second color data, in which the first color data comprises the first frequency components encoded therein, and the second color data comprises the second frequency components encoded therein; combining the first color data and the second color data, said combining yielding combined color data; utilizing one or more processors or electronic processing circuitry, detecting the encoded signal pattern from the combined color data, said detecting yielding rotation and scale information; and using the rotation and scale information to detect the encoded message from the combined color data. Of course, other combinations and claims are provided too.

An example method is described in which a processor receives a first plurality of symbols selected from a first set of symbols, translates the first plurality of symbols into a second plurality of symbols selected from an M-ary cyclic symbol set, and writes the second plurality of symbols to a surface. An additional example method is described in which a processor captures an image of at least a portion of a surface, detects a first plurality of symbols from the image, where the first plurality of symbols comprises symbols selected from an M-ary cyclic symbol set, translates the first plurality of symbols into a second plurality of symbols selected from a second set of symbols, and performs at least one task responsive to the second plurality of symbols.

In one embodiment, a command is received from a video provider at a media rendering device, the command being a command to embed a subscriber ID in at least one video frame in a video stream. A random key, k is also received from the video provider at the media rendering device. An injective transformation is invoked for k and the subscriber ID, the injective transformation generating a sequence of pairs of: near-transparent patterns and corresponding time periods. Logical blocks of the at least one video frame in the video stream are overlaid with one of the near-transparent patterns for its one of the corresponding time periods. Related apparatus, systems, and methods are also described.

In one example, a method for inserting a digital watermark in a signal includes obtaining the signal comprising a plurality of frames, inserting a first digital watermark in a first frame of the plurality of frames, inserting a second digital watermark in a second frame of the plurality of frames, wherein the second digital watermark differs from the first digital watermark in at least one way selected from a group of: a location within a respective frame, a number of bits, a pattern of bits, and a number of bits of a noise, and outputting a watermarked signal including the first digital watermark in the first frame and the second digital watermark in the second frame.

An optical code reader classifies a code depicted in input imagery, so that appropriate decoding actions can be invoked. This classification may identify, e.g., (a) whether the code is of a continuous tone or sparse mark variety, (b) which one of different reference signals it includes, and (c) which one of different protocols is used in expressing reference and payload signal components of the code. A great variety of other features and technologies are detailed as well.

Systems and methods are provided for decoding watermarks in video frames. A media device may receive a video frame that includes a first predetermined region comprising a watermark and a second predetermined region having pixel values selected to reduce the perceptibility of the first predetermined region. The media device may detect the watermark in the first predetermined region of the video frame and identify one or more contiguous subsets of pixels that correspond to a first pixel value and one or more contiguous subsets of pixels that correspond to a second pixel value. The media device then assigns a first symbol to the one or more contiguous subsets of pixels that correspond to the first pixel value and second symbol the one or more contiguous subsets of pixels that correspond to the second pixel value. The media device then generates a first sequence of symbols from the assigned symbols.

Data are encoded into one or more optically encoded images. The optically encoded images are then inserted as image data into a video sequence—i.e., in video frames. Data are transmitted in-band within the video, via any conceivable video distribution channel or format. The video may be trans-coded as required—because the data are optically encoded, any video processing that even crudely preserves the frame images will preserve the optically encoded data. This scheme of in-band data transfer in video is very robust. A video receiving apparatus receives the video, inspects the image data from video frames in memory, detects optically encoded images in the image data, and decodes the optically encoded images to recover the data. The frames carrying optically encoded images are typically discarded and not rendered to a display. The data from a plurality of optically encoded images may be concatenated, and further processed.

A method is provided for watermarking a machine learning model used for object detection or image classification. In the method, a first subset of a labeled set of ML training samples is selected. The first subset is of a predetermined class of images. In one embodiment, the first pixel pattern is selected and sized to have substantially the same dimensions as each sample of the first subset or each bounding box in the case of an object detector. Each sample of the first subset is relabeled to have a different label than the original label. An opacity of the pixel pattern may be adjusted independently for different parts of the pattern. The ML model is trained with the labeled set of ML training samples and the first subset of relabeled ML training samples. Using multiple different opacity factors provides both reliability and credibility to the watermark.

This patent document discloses physical documents including metameric ink pairs. One claim recites a document comprising: a first surface; a second surface, in which the first surface comprises a first set of print structures and a second set of print structures, in which the first set of print structures and the second set of print structures collective convey an encoded signal discernable from optical scan data representing at least a first portion of the first surface, in which the first set of print structures is provided on the first surface with a first ink and the second set of print structures is provided on the first surface with a second, different ink, and in which the first ink and the second, different ink comprise a metameric pair. Of course, other claims and combinations are described as well.

A method and computer programs for detecting document regions that will be excluded from a watermark embedding process are disclosed. The method comprises converting, by an adapter module, at least one page of a received document into a visual representation thereof, the visual representation keeping the position of the characters of the at least one page; receiving, by a text detector, the visual representation; processing, by the text detector, the visual representation using one or more artificial intelligence algorithms, and returning a list of invalid regions with their associated page positions as a result, wherein each invalid region of the list of invalid regions may have associated thereto a confidence score; and using, by a watermark embedding module or by a watermark extracting module, the list of invalid regions to provide a watermarked document or a message embedded in the document.