BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Perceptual Quality Assessment of NeRF and Neural View Synthesis Me
 thods for Front-Facing Views - Hanxue Liang\, Univ. of Cambridge
DTSTART:20240418T130000Z
DTEND:20240418T140000Z
UID:TALK215725@talks.cam.ac.uk
CONTACT:Rafal Mantiuk
DESCRIPTION:Neural view synthesis (NVS) is one of the most successful tech
 niques for synthesizing free viewpoint videos\, capable of achieving high 
 fidelity from only a sparse set of captured images. This success has led t
 o many variants of the techniques\, each evaluated on a set of test views 
 typically using image quality metrics such as PSNR\, SSIM\, or LPIPS. Ther
 e has been a lack of research on how NVS methods perform with respect to p
 erceived video quality. We present the first study on perceptual evaluatio
 n of NVS and NeRF variants. For this study\, we collected two datasets of 
 scenes captured in a controlled lab environment as well as in-the-wild. In
  contrast to existing datasets\, these scenes come with reference video se
 quences\, allowing us to test for temporal artifacts and subtle distortion
 s that are easily overlooked when viewing only static images. We measured 
 the quality of videos synthesized by several NVS methods in a well-control
 led perceptual quality assessment experiment as well as with many existing
 \nstate-of-the-art image/video quality metrics. We present a detailed anal
 ysis of the results and recommendations for dataset and metric selection f
 or NVS evaluation.\n\n[This is a practice talk for Eurographics 2024 prese
 ntation]. 
LOCATION:SS03 - William Gates Building
END:VEVENT
END:VCALENDAR