Eye and hand tracking utilizing lensless camera and machine learning
Eye and hand tracking systems in head-mounted display (HMD) devices are arranged with lensless camera systems using optical masks as encoding elements that apply convolutions to optical images of body parts (e.g., eyes or hands) of HMD device users. The convolved body images are scrambled or coded representations that are captured by a sensor in the system, but are not human-recognizable. A machine learning system such as a neural network is configured to extract body features directly from the coded representation without performance of deconvolutions conventionally utilized to reconstruct the original body images in human-recognizable form. The extracted body features are utilized by the respective eye or hand tracking systems to output relevant tracking data for the user's eyes or hands which may be utilized by the HMD device to support various applications and user experiences. The lensless camera and machine learning system are jointly optimizable on an end-to-end basis.
1. A body feature tracking system in a head-mounted display (HMD) device usable by a user, comprising:
an optical mask having a point spread function (PSF), the optical mask configured to convolve the PSF with reflected body features;
a sensor configured to record a convolution of the reflected body features and the PSF of the optical mask as a body feature map; and
a machine learning system configured to extract body features directly from the body feature map using a neural network while limiting performance of deconvolution to reconstruct an image of the reflected body features.
2. The body feature tracking system of claim 1 further comprising an illumination source configured to generate one of infrared (IR) light or visible light.
3. The body feature tracking system of claim 1 in which the sensor comprises one of a CMOS (complementary metal oxide semiconductor) image sensor or a CCD (charge-coupled device) image sensor.
4. The body feature tracking system of claim 1 in which the reflected features of the user's body comprise eye features, the sensor is configured to capture an eye feature map, and the machine learning system is configured to extract eye features from the eye feature map.
5. The body feature tracking system of claim 4 in which the machine learning system is further configured to estimate a gaze of the user based on the extracted eye features.
6. The body feature tracking system of claim 4 further comprising an illumination source configured to flood diffuse illumination to an eye of the user.
7. The body feature tracking system of claim 1 in which the reflected features of the user's body comprise hand features, the sensor is configured to capture a hand feature map, and the machine learning system is configured to extract hand features from the hand feature map.
8. The body feature tracking system of claim 7 in which the machine learning system is further configured to estimate one of location, orientation, or pose of a hand of the user based on the extracted hand features.