Avancerad sökning

Hittade 2 uppsatser som matchar ovanstående sökkriterier.

1. 1. A study of transfer learning on data-driven motion synthesis frameworks

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Nuo Chen; [2022]
   Nyckelord :Transfer learning; data-driven motion synthesis; objective-driven motion generation; rig-agnostic encoding; deep learning-based clustering model; procedural animation; Kunskapsöverföring; data-driven rörelsesyntetisering; procedurell animation; mål-driven animation-genereringsmodel; rig-agnostisk-kodning; djupinlärningsbaserad klusteringsmodel;

   Sammanfattning : Various research has shown the potential and robustness of deep learning-based approaches to synthesise novel motions of 3D characters in virtual environments, such as video games and films. The models are trained with the motion data that is bound to the respective character skeleton (rig). LÄS MER

3. 2. Self-supervised monocular depth estimation for dynamic scenes

   Master-uppsats, Lunds universitet/Matematik LTH

   Författare :Jonathan Lindberg; [2021]
   Nyckelord :Mathematics and Statistics;

   Sammanfattning : Estimating depth from an image is an ill-fitted problem, since we project a three-dimensional space to a two-dimensional image. However, human can estimate a plausible depth from only a single eye and often are we relaying on this for quick estimation of moving objects. LÄS MER