123. view synthesis with hierarchical clustering based occlusion filling

Department: Electrical & Computer Engineering
Faculty Advisor(s): Truong Nguyen

Primary Student
Name: Ji Dai
Email: jid046@ucsd.edu
Phone: 857-272-0251
Grad Year: 2020

Abstract
This paper presents a depth image based rendering algorithm for view synthesis task. We address the challenging occlusion filling problem with a hierarchical clustering approach. Depth distribution of neighboring pixels around each occlusion is explored and numbers of surrounding depth planes are determined with agglomerative clustering. Pixels from the most distant plane are picked as candidates to restore that occlusion. The proposed algorithm is evaluated on Middlebury stereo dataset and Microsoft Research 3D video dataset. Results show that our method ranks among the best performers.

Industry Application Area(s)
Software, Analytics

« Back to Posters or Search Results