Towards Accurate Facial Motion Retargeting with Identity-Consistent and Expression-Exclusive Constraints

Authors: Langyuan Mo, Haokun Li, Chaoyang Zou, Yubing Zhang, Ming Yang, Yihong Yang, Mingkui Tan1981-1989

AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on facial motion retargeting and 3D face reconstruction tasks demonstrate the superiority of the proposed method over existing methods. Our code and supplementary materials are available at https://github.com/deepmo24/CPEM.
Researcher Affiliation Collaboration Langyuan Mo1,2, Haokun Li1, Chaoyang Zou3, Yubing Zhang3, Ming Yang3, Yihong Yang4, Mingkui Tan1,5* 1 School of Software Engineering, South China University of Technology, 2 Pazhou Laboratory, 3 CVTE Research, 4 MINIEYE, 5 Key Laboratory of Big Data and Intelligent Robot (South China University of Technology), Ministry of Education
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code Yes Our code and supplementary materials are available at https://github.com/deepmo24/CPEM.
Open Datasets Yes We train our model on three publicly available datasets: Vox Celeb2 (Joon Son et al. 2018), 300W-LP (Zhu et al. 2016) and FEAFA (Yan et al. 2019).
Dataset Splits No The paper mentions training and testing but does not explicitly detail validation splits or provide explicit split percentages for any dataset.
Hardware Specification No The paper mentions using ResNet50 as the backbone but does not specify any hardware details like GPU model, CPU type, or memory.
Software Dependencies Yes We implement our method based on Py Torch (Paszke et al. 2019) and use the differentiable renderer from Pytorch3d (Lassner and Zollhofer 2021).
Experiment Setup Yes We use an Adam optimizer (Kingma and Ba 2015) with a learning rate of 1e-4. We train our model for 300K iterations with a batch size of 8 and an input size of 224x224, and only use the expression-exclusive loss in the last 100K iterations. ... By default, we set T = 4, λidc = 1000, and λexp = 10.