peterzhousz / estimating_mechanical_properties_of_cloth Goto Github PK

View Code? Open in Web Editor NEW

Estimating mechanical properties of cloth from videos using dense motion trajectories: human psychophysics and machine learning

MATLAB 18.21% R 0.12% Shell 0.33% Makefile 2.66% HTML 5.57% CSS 1.15% Python 4.54% Clean 0.24% TeX 0.24% C 64.43% Objective-C 0.99% Roff 0.39% C++ 1.01% Mercury 0.11% M 0.02%

estimating_mechanical_properties_of_cloth's Introduction

Author: Wenyan Bi

since 2019-10-13

Estimating mechanical properties of cloth from videos using dense motion trajectories: Human psychophysics and machine learning

Description

In this project, we use Blender (2.7.6) rendered cloth animations as our dataset.

First, we used a maximum likelihood differential scaling (MLDS) method to measure the human perceptual scale of cloth stiffness. Codes and instructions of this experiment can be found in the MLDS_Experiment folder.

Maximum likelihood differential scaling (MLDS)

Next, we extracted the dense motion trajectory features of all the cloth videos.

Dense motion trajectory features

Using the extracted dense motion trajectory features, we built a support vector regression (SVR) model to predict the human perceptual scale of stiffness. Codes and instruction of this experiment can be found in the MotionAnalysis folder.

Computational modeling of human perceptual scale

Dependencies

Psychtoolbox in Matlab.
VLFeat Matlab toolbox (vlfeat-0.9.20).
Yael for Matlab (yael_v438).
Dense trajectories video descriptors (third version).
ffmpeg-0.11.1.

References

If you use the codes, please cite the following papers.

1. Bi, W., Jin, P., Nienborg, H., & Xiao, B. (2018). Estimating mechanical properties of cloth from videos using dense motion trajectories: Human psychophysics and machine learning. Journal of Vision, 18(5):12, 1–20.
2. Brainard, D. H., & Vision, S. (1997). The psychophysics toolbox. Spatial vision, 10, 433-436.
3. Knoblauch, K., & Maloney, L. T. (2008). MLDS: Maximum likelihood difference scaling in R. Journal of Statistical Software, 25(2), 1–26.
4. Maloney, L. T., & Yang, J. N. (2003). Maximum likelihood difference scaling. Journal of Vision, 3(8): 5, 573–585.
5. Wang, H., Kläser, A., Schmid, C., & Cheng-Lin, L. (2011, June). Action recognition by dense trajectories.
6. Wang, H., & Schmid, C. (2013). Action recognition with improved trajectories. In Proceedings of the IEEE international conference on computer vision (pp. 3551-3558).