Cross-Species 3D Face Morphing via Alignment-Aware Controller

Authors: Xirui Yan, Zhenbo Yu, Bingbing Ni, Hang Wang3018-3026

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experimental results in multiple people and animals demonstrate that our method produces high-quality deformation results. ... Experiments ... Evaluation Metrics ... Morphing Results ... Ablation Studies
Researcher Affiliation Academia Shanghai Jiao Tong University, Shanghai 200240, China
Pseudocode No The paper does not contain any pseudocode or clearly labeled algorithm blocks.
Open Source Code No The paper does not provide any statement or link indicating the availability of open-source code for the described methodology.
Open Datasets Yes COMA (Ranjan et al. 2018) is a 3D dataset which contains 20466 head meshes of 12 different subjects. The Headspace Dataset (Dai et al. 2020) contains 3D human faces of 1519 people with detailed BMP textures.
Dataset Splits No The paper describes the training and test sets but does not specify a separate validation dataset split.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments.
Software Dependencies No The paper mentions that network layers are 'the same as encoders and the decoders used in Neural Cages (Wang et al. 2020) which are the simplified versions of Atlas Net (Groueix et al. 2018)' but does not list specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup No The paper states: 'In total, our training loss is formulated as: L = LEC + Ldeform + LCBM + Leye + Lsmooth, where we have omitted a set of hyper-parameters that balance the importance of different loss terms here, and provide detailed descriptions in the supplementary materials.' This indicates specific experimental setup details (hyperparameters) are not in the main text.