D ShinFollow


Many video conferencing applications support portrait mode where the participant is in focus and their surroundings are blurred. Portrait mode based on computer vision techniques can sometimes be unsatisfactory due to difficulties in separating the user and the background. This disclosure describes techniques that use existing ultra-wideband (UWB) hardware on commodity devices to perform crisp and more accurate segmentation. A UWB handshake protocol is utilized to estimate the distance and angle between the camera and another device that the conference participant is wearing. The estimate is used to automatically adjust camera focal length to focus on the conference participant while blurring other objects. The techniques can make high-resolution portrait mode video conferencing affordable for users with commodity devices.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.