CtrlAvatar: Controllable Avatars Generation via Disentangled Invertible Networks
Authors: Wenfeng Song, Yang Ding, Fei Hou, Shuai Li, Aimin Hao, Xia Hou
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we present an overview of the datasets and evaluation metrics. We conduct both quantitative and qualitative comparisons of our method against the state-of-the-art in terms of reconstruction accuracy, texture quality, and inference time. Subsequently, we conduct ablation studies to confirm the efficacy of our critical design choices within key modules. Additionally, we show the result of driving and editing the human avatar. |
| Researcher Affiliation | Academia | Wenfeng Song1, Yang Ding1, Fei Hou2,3, Shuai Li4,5*, Aimin Hao4, Xia Hou1 1College of Computer Science, Beijing Information Science and Technology University 2Key Laboratory of System Software (CAS), State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences, China 3University of Chinese Academy of Sciences, China 4State Key Laboratory of Virtual Reality Technology and Systems, Beihang University 5Zhongguancun Laboratory, China EMAIL, EMAIL, EMAIL, EMAIL |
| Pseudocode | No | The paper describes its methodology using mathematical equations and textual explanations, but it does not include any clearly labeled pseudocode blocks or algorithms. |
| Open Source Code | Yes | Code https://github.com/1211186431/Ctrl Avatar |
| Open Datasets | Yes | SX-Humans. X-Humans (Shen et al. 2023) is a comprehensive 3D clothing scanning dataset featuring textured human body scans. It includes 20 subjects, each with continuous scanning action sequences. To evaluate our method with limited data, we curated a smaller subset, SX-Humans, by selecting four scanning actions from each subject s sequences at predetermined intervals. S-Custom Humans. Custom Humans (Ho et al. 2023) is a dataset comprising over 600 high-quality scans from a volumetric capture of 80 participants in 120 different garments and poses. |
| Dataset Splits | No | The paper mentions curating subsets (SX-Humans, S-Custom Humans) from existing datasets, but it does not provide specific details on how these subsets are further divided into training, validation, or test splits (e.g., percentages, sample counts, or methodology for splitting). |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU models, CPU types, or memory used for running the experiments. |
| Software Dependencies | No | The paper does not provide specific software dependency details with version numbers (e.g., Python, PyTorch, CUDA versions) needed to replicate the experiments. |
| Experiment Setup | No | The paper states, "Further experimental details are provided in the Supplementary Material." However, the main text itself does not contain specific experimental setup details such as hyperparameter values, learning rates, batch sizes, or optimizer settings. |