Temporally coherent mesh sequence segmentations

Romain Arcila 1, 2 Cédric Cagniart 3 Franck Hétroy 2 Edmond Boyer 2 Florent Dupont 1
1 M2DisCo - Geometry Processing and Constrained Optimization
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
2 MORPHEO - Capture and Analysis of Shapes in Motion
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : In this report, we consider the problem of fully automatic segmentation of mesh sequences, with or without temporal coherence. More precisely, our goal is to identify model parts that consistently move rigidly over time. We propose a novel framework that incrementally adapts segments along a sequence based on the coherence of motion information within each segment. In contrast to existing approaches, this framework handles meshes independently reconstructed at each time instant, provided that motion cues are available. It allows therefore for meshes with varying connectivity as well as varying topology. Experiments on various data sets in addition to a quantitative evaluation demonstrate the effectiveness and robustness of the approach.
Complete list of metadatas


https://hal.inria.fr/hal-00658060
Contributor : Franck Hétroy-Wheeler <>
Submitted on : Tuesday, January 10, 2012 - 9:47:41 AM
Last modification on : Tuesday, June 25, 2019 - 1:27:22 AM
Long-term archiving on : Tuesday, December 13, 2016 - 9:19:49 PM

Files

RR-7856.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00658060, version 2

Citation

Romain Arcila, Cédric Cagniart, Franck Hétroy, Edmond Boyer, Florent Dupont. Temporally coherent mesh sequence segmentations. [Research Report] RR-7856, INRIA. 2012, pp.20. ⟨hal-00658060v2⟩

Share

Metrics

Record views

752

Files downloads

948