The Effects of Motion Blur and Frame Rate on Perceived Cinematographic Realism
Author
Olsson, Anna Bünning
Term
4. term
Education
Publication year
2013
Submitted on
2013-05-24
Pages
51
Abstract
Dette speciale undersøger, hvordan billedfrekvens (antal billeder pr. sekund) og bevægelsessløring (den udtværing, der opstår ved hurtige bevægelser i kamerabilleder og ofte simuleres i computergrafik) påvirker den oplevede filmiske realisme i en computeranimeret actionscene, altså hvor filmisk og overbevisende scenen føles for seere. Specialet drøfter også, hvad realisme betyder i computergrafik, og hvorfor publikum accepterer visuelle effekter som en del af det, de ser, selv når sund fornuft siger, at noget er umuligt. Gennemgangen af visuel perception, bevægelsessløring, billedfrekvens og realisme munder ud i testbare hypoteser. Herefter gennemføres et eksperiment med to betingelser: sekvenser med forskellige billedfrekvenser med bevægelsessløring og sekvenser med forskellige billedfrekvenser uden bevægelsessløring. Resultaterne sammenlignes både inden for og på tværs af betingelser. På grund af fejl i testopstillingen blev fundene uklare, men de indsamlede data tyder på, at mere entydige resultater kunne være opnået uden disse fejl.
This thesis investigates how frame rate (the number of images shown per second) and motion blur (the streaking that occurs during fast movement in camera footage and is often simulated in computer graphics) shape the perceived cinematographic realism of a computer-animated action scene, that is, how film-like and convincing it feels to viewers. It also considers what realism means in computer graphics and why audiences accept visual effects as part of what they see, even when common sense suggests some events are impossible. A review of visual perception, motion blur, frame rate, and realism leads to testable hypotheses. An experiment then compares sequences at different frame rates in two conditions: with motion blur and without motion blur. Results are analyzed within and across conditions. Due to errors in the test setup, findings were inconclusive, but the collected data indicate that clearer patterns might have emerged without these errors.
[This abstract was generated with the help of AI]
Documents