Abstract

We present a novel hybrid representation for character animation from 4D Performance Capture (4DPC) data which combines skeletal control with surface motion graphs. 4DPC data are temporally aligned 3D mesh sequence reconstructions of the dynamic surface shape and associated appearance from multiple-view video. The hybrid representation supports the production of novel surface sequences which satisfy constraints from user-specified key-frames or a target skeletal motion. Motion graph path optimisation concatenates fragments of 4DPC data to satisfy the constraints while maintaining plausible surface motion at transitions between sequences. Space-time editing of the mesh sequence using a learned part-based Laplacian surface deformation model is performed to match the target skeletal motion and transition between sequences. The approach is quantitatively evaluated for three 4DPC datasets with a variety of clothing styles. Results for key-frame animation demonstrate production of novel sequences that satisfy constraints on timing and position of less than 1% of the sequence duration and path length. Evaluation of motion-capture-driven animation over a corpus of 130 sequences shows that the synthesised motion accurately matches the target skeletal motion. The combination of skeletal control with the surface motion graph extends the range and style of motion which can be produced while maintaining the natural dynamics of shape and appearance from the captured performance.

Paper

Hybrid Skeletal-Surface Motion Graphs for Character Animation from 4D Performance Capture
Peng Huang, Margara Tejera, John Collomosse and Adrian Hilton
ACM Transactions on Graphics (TOG)
Volume 34 Issue 2
February 2015
Article No. 17




WebGL Demo

Note: WebGL enabled web browser is required, e.g. lasted version chrome or firefox

Demo 1 - Interactive 3D Game Character

Demo 2 - Interactive 3D Fashion Character

Data

4D video datasets used in this paper is available for download at CVSSP 3D Data Repository.

Please send an email to to obtain the password to download the data.

Dataset Character1

Hit ‒ 76 frames ‒ 8 cameras
WebGL preview
Jog ‒ 26 frames ‒ 8 cameras
WebGL preview
Jog2Walk ‒ 28 frames ‒ 8 cameras
WebGL preview
Stagger ‒ 81 frames ‒ 8 cameras
WebGL preview
Stand ‒ 50 frames ‒ 8 cameras
WebGL preview
Stand2Walk ‒ 31 frames ‒ 8 cameras
WebGL preview
Tense ‒ 35 frames ‒ 8 cameras
WebGL preview
Walk ‒ 31 frames ‒ 8 cameras
WebGL preview
Walk2Jog ‒ 27 frames ‒ 8 cameras
WebGL preview
Walk2Stand ‒ 56 frames ‒ 8 cameras
WebGL preview

Dataset Fashion1

Pose ‒ 125 frames ‒ 8 cameras
WebGL preview
Stand2Walk ‒ 46 frames ‒ 8 cameras
WebGL preview
Twirl ‒ 101 frames ‒ 8 cameras
WebGL preview
Walk ‒ 28 frames ‒ 8 cameras
WebGL preview
Walk2Stand ‒ 57 frames ‒ 8 cameras
WebGL preview
WalkPose ‒ 72 frames ‒ 8 cameras
WebGL preview

Acknowledgments

This research was supported by EU FP7 Project RE@CT: Immersive production and delivery of interactive 3D content and EPSRC/BBC industrial case studentship award ‘3D Video Editing’.

Related Works