XLXiang Liu
Xiang Liu
*This is an algorithmically generated author record based on publicly available information.
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiang Liu has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Shiguo Lian
Data Science & Artificial Intelligence Research Institute, China Unicom, No. 21 Financial Street, Beijing 100013, China
8 shared publications - Zhaoxiang Liu
China Unicom
8 shared publications - Huan Hu
China Unicom
7 shared publications - Kai Wang
University of Pennsylvania
7 shared publications - Zezhou Chen
China United Network Communications Group (Beijing, China)
4 shared publications - Zipeng Wang
Unicom Data Intelligence, China Unicom, Beijing, China and Data Science & Artificial Intelligence Research Institute, China Unicom, Beijing, China
4 shared publications
Latest Publications
- 1 November 2025
- Published by Springer Nature
- p. 89-101
Art3D-Fusion: A Hybrid Framework for Visual Synthesis with Artistic Control
- 1 November 2025
- Published by Springer Nature
- p. 89-101
Art3D-Fusion: A Hybrid Framework for Visual Synthesis with Artistic Control
- Open Access
PSTF-AttControl: Per-Subject-Tuning-Free Personalized Image Generation with Controllable Face Attributes
Recent advancements in personalized image generation have significantly improved facial identity preservation, particularly in fields such as entertainment and social media. However, existing methods still struggle to achieve precise control over facial attributes in a per-subject-tuning-free (PSTF) way. Tuning-based techniques like PreciseControl have shown promise by providing fine-grained control over facial features, but they often require extensive technical expertise and additional training data, limiting their accessibility. In contrast, PSTF approaches simplify the process by enabling image generation from a single facial input, but they lack precise control over facial attributes. In this paper, we introduce a novel, PSTF method that enables both precise control over facial attributes and high-fidelity preservation of facial identity. Our approach utilizes a face recognition model to extract facial identity features, which are then mapped into the $W^+$ latent space of StyleGAN2 using the e4e encoder. We further enhance the model with a Triplet-Decoupled Cross-Attention module, which integrates facial identity, attribute features, and text embeddings into the UNet architecture, ensuring clean separation of identity and attribute information. Trained on the FFHQ dataset, our method allows for the generation of personalized images with fine-grained control over facial attributes, while without requiring additional fine-tuning or training data for individual identities. We demonstrate that our approach successfully balances personalization with precise facial attribute control, offering a more efficient and user-friendly solution for high-quality, adaptable facial image synthesis. The code is publicly available at https://github.com/UnicomAI/PSTF-AttControl.Open AccessPSTF-AttControl: Per-Subject-Tuning-Free Personalized Image Generation with Controllable Face Attributes
Recent advancements in personalized image generation have significantly improved facial identity preservation, particularly in fields such as entertainment and social media. However, existing methods still struggle to achieve precise control over facial attributes in a per-subject-tuning-free (PSTF) way. Tuning-based techniques like PreciseControl have shown promise by providing fine-grained control over facial features, but they often require extensive technical expertise and additional training data, limiting their accessibility. In contrast, PSTF approaches simplify the process by enabling image generation from a single facial input, but they lack precise control over facial attributes. In this paper, we introduce a novel, PSTF method that enables both precise control over facial attributes and high-fidelity preservation of facial identity. Our approach utilizes a face recognition model to extract facial identity features, which are then mapped into the $W^+$ latent space of StyleGAN2 using the e4e encoder. We further enhance the model with a Triplet-Decoupled Cross-Attention module, which integrates facial identity, attribute features, and text embeddings into the UNet architecture, ensuring clean separation of identity and attribute information. Trained on the FFHQ dataset, our method allows for the generation of personalized images with fine-grained control over facial attributes, while without requiring additional fine-tuning or training data for individual identities. We demonstrate that our approach successfully balances personalization with precise facial attribute control, offering a more efficient and user-friendly solution for high-quality, adaptable facial image synthesis. The code is publicly available at https://github.com/UnicomAI/PSTF-AttControl. - Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9724-9732
CP3: Customizable 3D Pop-Out Effect Creation for Immersive Content Using Multimodal Models
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9724-9732
Open AccessCP3: Customizable 3D Pop-Out Effect Creation for Immersive Content Using Multimodal Models
- 16 October 2025
- Published by Elsevier in Image and Vision Computing
- Vol. 164
PSTF-AttControl: Per-subject-tuning-free personalized image generation with controllable face attributes
- 16 October 2025
- Published by Elsevier in Image and Vision Computing
- Vol. 164
PSTF-AttControl: Per-subject-tuning-free personalized image generation with controllable face attributes
- 25 July 2025
- Published by Springer Nature
- p. 125-139
ILearnRobot: An Interactive Learning-Based Multi-modal Robot with Continuous Improvement
- 25 July 2025
- Published by Springer Nature
- p. 125-139
ILearnRobot: An Interactive Learning-Based Multi-modal Robot with Continuous Improvement
- 22 July 2025
- Published by Springer Nature
- p. 250-260
SLearnLLM: A Self-Learning Framework for Efficient Domain-Specific Adaptation of Large Language Models
- 22 July 2025
- Published by Springer Nature
- p. 250-260
SLearnLLM: A Self-Learning Framework for Efficient Domain-Specific Adaptation of Large Language Models
- 26 October 2024
- Published by Springer Nature
- p. 157-170
A Multimodal Benchmark Dataset and Model for Crop Disease Diagnosis
- 26 October 2024
- Published by Springer Nature
- p. 157-170
A Multimodal Benchmark Dataset and Model for Crop Disease Diagnosis
- 31 July 2024
- Published by Springer Nature
- p. 114-130
Spatial-Temporal Transformer Network for Continuous Action Recognition in Industrial Assembly
- 31 July 2024
- Published by Springer Nature
- p. 114-130
Spatial-Temporal Transformer Network for Continuous Action Recognition in Industrial Assembly