Motion-Compensated Viewpoint Shift

Eye contact is an essential social cue that conveys our attention to others but is difficult to maintain during video calls. Many existing methods to synthesize a gaze-corrected view involve estimating a 3D face model and projecting it into the desired camera view, which is too computationally expen...

Full description

Bibliographic Details
Main Author: Tao, Julius L.
Other Authors: Lee, Chong U.
Format: Thesis
Published: Massachusetts Institute of Technology 2024
Online Access:https://hdl.handle.net/1721.1/156638
_version_ 1811074436222156800
author Tao, Julius L.
author2 Lee, Chong U.
author_facet Lee, Chong U.
Tao, Julius L.
author_sort Tao, Julius L.
collection MIT
description Eye contact is an essential social cue that conveys our attention to others but is difficult to maintain during video calls. Many existing methods to synthesize a gaze-corrected view involve estimating a 3D face model and projecting it into the desired camera view, which is too computationally expensive for most personal computers. By drawing inspiration from 2D methods of video frame interpolation, we wish to not only correct eye gaze but also better align the face towards the camera without this expensive 3D modeling. Our findings suggest that adding a second webcam opposite the first and interpolating between the two outer camera views can give realistic, gaze-aligned center views. We conclude that the prevailing approach of 3D modeling is surprisingly not necessary for gaze correction. Not only do 2D techniques suffice, but their synthesized frames can appear more natural than prior results. We believe that this work is a crucial step towards true-to-life viewpoint shift for live video conferences.
first_indexed 2024-09-23T09:49:26Z
format Thesis
id mit-1721.1/156638
institution Massachusetts Institute of Technology
last_indexed 2024-09-23T09:49:26Z
publishDate 2024
publisher Massachusetts Institute of Technology
record_format dspace
spelling mit-1721.1/1566382024-09-04T03:51:45Z Motion-Compensated Viewpoint Shift Tao, Julius L. Lee, Chong U. Lim, Jae S. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Eye contact is an essential social cue that conveys our attention to others but is difficult to maintain during video calls. Many existing methods to synthesize a gaze-corrected view involve estimating a 3D face model and projecting it into the desired camera view, which is too computationally expensive for most personal computers. By drawing inspiration from 2D methods of video frame interpolation, we wish to not only correct eye gaze but also better align the face towards the camera without this expensive 3D modeling. Our findings suggest that adding a second webcam opposite the first and interpolating between the two outer camera views can give realistic, gaze-aligned center views. We conclude that the prevailing approach of 3D modeling is surprisingly not necessary for gaze correction. Not only do 2D techniques suffice, but their synthesized frames can appear more natural than prior results. We believe that this work is a crucial step towards true-to-life viewpoint shift for live video conferences. M.Eng. 2024-09-03T21:13:49Z 2024-09-03T21:13:49Z 2024-05 2024-07-11T14:36:17.388Z Thesis https://hdl.handle.net/1721.1/156638 In Copyright - Educational Use Permitted Copyright retained by author(s) https://rightsstatements.org/page/InC-EDU/1.0/ application/pdf Massachusetts Institute of Technology
spellingShingle Tao, Julius L.
Motion-Compensated Viewpoint Shift
title Motion-Compensated Viewpoint Shift
title_full Motion-Compensated Viewpoint Shift
title_fullStr Motion-Compensated Viewpoint Shift
title_full_unstemmed Motion-Compensated Viewpoint Shift
title_short Motion-Compensated Viewpoint Shift
title_sort motion compensated viewpoint shift
url https://hdl.handle.net/1721.1/156638
work_keys_str_mv AT taojuliusl motioncompensatedviewpointshift