Abstract: Techniques for evaluating images are disclosed. An image is displayed. Data corresponding to a gaze fixation of a user who is viewing the image is obtained. An image artifact is identified based on the gaze fixation. The image artifact is at a location in the image where the gaze fixation occurred. A recording of the user's utterance is accessed. This recording is recorded during an overlapping time period with when the gaze fixation occurred. The recording is transcribed into text, which is then parsed. A key term is extracted from the parsed text. A determination is made as to whether the key term corresponds to the image artifact identified based on the gaze fixation.

Abstract: The disclosed techniques are focused on processes for encoding an enhanced image with non-image data. Notably, the “non-image data” is distinct from “image data” in that the image data defines display characteristics of an image (e.g., display properties of a pixel) while the non-image data is unrestricted and can describe any data, even data different than display characteristics. An image is accessed, where the image includes at least one pixel that is associated with at least one color channel. Non-image data is encoded into the color channel. An index, which maps where the non-image data has been encoded in the color channel of the pixel, is generated or modified. As a result of encoding the non-image data into the color channel, an enhanced image is generated.

Abstract: Techniques are disclosed for aligning fiducial markers that commonly exist in each of multiple different N-dimensional (N-D) data sets. Notably, the N-D data sets are at least three-dimensional (3D) data sets. A first set and a second set of N-D data are accessed. A set of one or more fiducial markers that commonly exist in both those sets are identified. Based on the fiducial markers, one or more transformations are performed to align the two sets. Performing this alignment process results in at least a selected number of the common fiducial markers that exist in the two sets being within a threshold alignment relative to one another.

Abstract: Techniques are disclosed for aligning fiducial markers that commonly exist in each of multiple different N-dimensional (N-D) data sets. Notably, the N-D data sets are at least three-dimensional (3D) data sets. A first set and a second set of N-D data are accessed. A set of one or more fiducial markers that commonly exist in both those sets are identified. Based on the fiducial markers, one or more transformations are performed to align the two sets. Performing this alignment process results in at least a selected number of the common fiducial markers that exist in the two sets being within a threshold alignment relative to one another.

Abstract: The disclosed techniques are focused on processes for encoding an enhanced image with non-image data. Notably, the “non-image data” is distinct from “image data” in that the image data defines display characteristics of an image (e.g., display properties of a pixel) while the non-image data is unrestricted and can describe any data, even data different than display characteristics. An image is accessed, where the image includes at least one pixel that is associated with at least one color channel. Non-image data is encoded into the color channel. An index, which maps where the non-image data has been encoded in the color channel of the pixel, is generated or modified. As a result of encoding the non-image data into the color channel, an enhanced image is generated.