Generating personalized end user head-related transfer function (HRTV) using panoramic images of ear
AssigneeSony Corporation
InventorsJames R. Milne, Gregory Peter Carlsson
Life of this patent
Estimating the term…
Abstract
Left and right ear HRTF coefficients are determined for an end user, one each for each of a plurality of head orientations, and provided to the end user on a portable recording medium, or via the Internet, etc. The user can then implement the files on audio played on the user's headphones, with the file corresponding to the user's head orientation being selected as the user moves his head to ensure the sound as perceived by the user remains emanating from a fixed external location. The user's personal HRTF may be generated using panoramic images of the ears, such as multiple 2D images or 3D imaging.
Claims20
An apparatus comprising:
- at least one processor configured with instructions to:
- receive at least first and second photographs of a person's ear from at least first and second camera positions;
- based at least in part on the at least first and second photographs, generate a head-related transfer function (HRTF) to be applied to audio for the person's ear;
- and
- present input option selectable by the person to select a HRTF type from a group of types comprising one or more of stereo, multi-channel, up-mix from stereo to multichannel.
A method, comprising:
- receiving at least one photograph of a region of a person's ear, the at least one photograph comprising at least one three-dimensional (3D) image or at least two two-dimensional (2D) images from at least first and second camera positions;
- based at least in part on the at one photograph, generating a head-related transfer function (HRTF) to be applied to audio for the person's ear;
- and
- combining the HRTF with information pertaining to a virtual venue in which to simulate playing audio.
An apparatus comprising:
- at least one processor
- operable to access at least one database associating images of ears with respective head-related transfer functions (HRTFs);
- and
- wherein the at least one processor is programmed with instructions to:
- identify in at least one database a closest match to at least one image of a person;
- return at least one HRTF associated with the closest match;
- and
- play back audio using fewer than all taps of the HRTF.
Dependencies are inferred from the claim text.