Yang Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yang Liu has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
Xueming Li
Beijing University of Posts and Telecommunications
5 shared publications- Xueming Li
Beijing University of Posts and Telecommunications, Beijing, China
4 shared publications Fangyu Li
Beijing University of Posts and Telecommunications,School of Digital Media and Design Art
3 shared publicationsJiaqi Tan
Beijing University of Posts and Telecommunications,School of Digital Media and Design Art
3 shared publications- Jiaxin Tong
School of Digtal Media and Design Art, Beijing University of Posts and Telecommunications, Beijing, Beijing, China
3 shared publications - Mingtong Zhang
School of Digtal Media and Design Art, Beijing University of Posts and Telecommunications, Beijing, Beijing, China
3 shared publications
Latest Publications
- 3 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 8912-8916
Instruction Guided Multi Object Image Editing with Quantity and Layout Consistency
Instruction driven image editing with standard CLIP text encoders often fails in complex scenes with many objects. We present QL-Adapter, a framework for multiple object editing that tackles two challenges: enforcing object counts and spatial layouts, and accommodating diverse categories. QL-Adapter consists of two core modules: the Image-Layout Fusion Module (ILFM) and the Cross-Modal Augmentation Module (CMAM). ILFM fuses layout priors with ViT patch tokens from the CLIP image encoder to strengthen spatial structure understanding. CMAM injects image features into the text branch to enrich textual embeddings and improve instruction following. We further build QL-Dataset, a benchmark that spans broad category, layout, and count variations, and define the task of quantity and layout consistent image editing (QL-Edit). Extensive experiments show that QL-Adapter achieves state of the art performance on QL-Edit and significantly outperforms existing models.
- 3 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 8912-8916
Instruction Guided Multi Object Image Editing with Quantity and Layout Consistency
Instruction driven image editing with standard CLIP text encoders often fails in complex scenes with many objects. We present QL-Adapter, a framework for multiple object editing that tackles two challenges: enforcing object counts and spatial layouts, and accommodating diverse categories. QL-Adapter consists of two core modules: the Image-Layout Fusion Module (ILFM) and the Cross-Modal Augmentation Module (CMAM). ILFM fuses layout priors with ViT patch tokens from the CLIP image encoder to strengthen spatial structure understanding. CMAM injects image features into the text branch to enrich textual embeddings and improve instruction following. We further build QL-Dataset, a benchmark that spans broad category, layout, and count variations, and define the task of quantity and layout consistent image editing (QL-Edit). Extensive experiments show that QL-Adapter achieves state of the art performance on QL-Edit and significantly outperforms existing models.
- Open Access
Instruction Guided Multi Object Image Editing with Quantity and Layout Consistency
Instruction driven image editing with standard CLIP text encoders often fails in complex scenes with many objects. We present QL-Adapter, a framework for multiple object editing that tackles two challenges: enforcing object counts and spatial layouts, and accommodating diverse categories. QL-Adapter consists of two core modules: the Image-Layout Fusion Module (ILFM) and the Cross-Modal Augmentation Module (CMAM). ILFM fuses layout priors with ViT patch tokens from the CLIP image encoder to strengthen spatial structure understanding. CMAM injects image features into the text branch to enrich textual embeddings and improve instruction following. We further build QL-Dataset, a benchmark that spans broad category, layout, and count variations, and define the task of quantity and layout consistent image editing (QL-Edit). Extensive experiments show that QL-Adapter achieves state of the art performance on QL-Edit and significantly outperforms existing models.
Open AccessInstruction Guided Multi Object Image Editing with Quantity and Layout Consistency
Instruction driven image editing with standard CLIP text encoders often fails in complex scenes with many objects. We present QL-Adapter, a framework for multiple object editing that tackles two challenges: enforcing object counts and spatial layouts, and accommodating diverse categories. QL-Adapter consists of two core modules: the Image-Layout Fusion Module (ILFM) and the Cross-Modal Augmentation Module (CMAM). ILFM fuses layout priors with ViT patch tokens from the CLIP image encoder to strengthen spatial structure understanding. CMAM injects image features into the text branch to enrich textual embeddings and improve instruction following. We further build QL-Dataset, a benchmark that spans broad category, layout, and count variations, and define the task of quantity and layout consistent image editing (QL-Edit). Extensive experiments show that QL-Adapter achieves state of the art performance on QL-Edit and significantly outperforms existing models.
- Open Access
Robust Multimodal Semantic Segmentation with Balanced Modality Contributions
Multimodal semantic segmentation enhances model robustness by exploiting cross-modal complementarities. However, existing methods often suffer from imbalanced modal dependencies, where overall performance degrades significantly once a dominant modality deteriorates in real-world scenarios. Thus, modality balance has become acritical challenge for practical multimodal segmentation. To address this issue, we propose EQUISeg, a multimodal segmentation framework that balances modality contributions through equal encoding of modalities. Built upon a four-stage Cross-modal Transformer Block(CMTB), EQUISeg enables efficient multimodal fusion and hierarchical selection. Furthermore, we design a Self-guided Module(SGM) that mitigates modality imbalance by introducing a mutual guidance mechanism, enabling each modality to adaptively adjust its contribution and enhance robustness under degraded conditions. Extensive experiments on multiple datasets demonstrate that EQUISeg achieves significant performance gains and effectively alleviates the adverse effects of modality imbalance in segmentation tasks.
Open AccessRobust Multimodal Semantic Segmentation with Balanced Modality Contributions
Multimodal semantic segmentation enhances model robustness by exploiting cross-modal complementarities. However, existing methods often suffer from imbalanced modal dependencies, where overall performance degrades significantly once a dominant modality deteriorates in real-world scenarios. Thus, modality balance has become acritical challenge for practical multimodal segmentation. To address this issue, we propose EQUISeg, a multimodal segmentation framework that balances modality contributions through equal encoding of modalities. Built upon a four-stage Cross-modal Transformer Block(CMTB), EQUISeg enables efficient multimodal fusion and hierarchical selection. Furthermore, we design a Self-guided Module(SGM) that mitigates modality imbalance by introducing a mutual guidance mechanism, enabling each modality to adaptively adjust its contribution and enhance robustness under degraded conditions. Extensive experiments on multiple datasets demonstrate that EQUISeg achieves significant performance gains and effectively alleviates the adverse effects of modality imbalance in segmentation tasks.
- Open Access
- 21 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 441-446
Optimization of Albedo Map in 3D Face Reconstruction
- 21 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 441-446
Open AccessOptimization of Albedo Map in 3D Face Reconstruction
- Open Access
- 21 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 728-735
Language-guided Human Motion Generation with Temporal Contact Diffusion in 3D Scenes
- 21 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 728-735
Open AccessLanguage-guided Human Motion Generation with Temporal Contact Diffusion in 3D Scenes
- Open Access
- 21 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 428-434
TailorCode: 3D Clothed Human Generation from Images with Text-Instructed Editable Clothing Dimensions
- 21 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 428-434
Open AccessTailorCode: 3D Clothed Human Generation from Images with Text-Instructed Editable Clothing Dimensions
- Open Access
- 14 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 199-205
ACoSkeNet: A unique automatic coloring of sketches model based on U-Net
In recent years, sketches have attracted much attention as a form of free media, and at the same time, the automatic coloring task of hand-drawn sketches has become one of the research hotspots in the field of vision. Sketch coloring is a challenging task, mainly due to: a) lack of inherent color information; b) the result of coloring is prone to cross-color and overflow; c) poor controllability of the coloring process. Based on this, we design a unique deep network that combines self-attention mechanism to alleviate this problem in this paper. Specifically, we first use the encoding and decoding structure of U-net to capture the delicate details in the sketch; a unique self-attention module is then integrated to automatically capture key areas of the input sketch, thereby improving color distribution modeling and contextual awareness, enhancing the results of automatic coloring. Experiments on a large-scale anime sketch dataset show that the proposed method exhibits significant performance improvement, capturing delicate details in sketches while maintaining faithful color restoration. In addition, our model also supports guided coloring based on provided clues, which providing users more controllability.
- 14 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 199-205
Open AccessACoSkeNet: A unique automatic coloring of sketches model based on U-Net
In recent years, sketches have attracted much attention as a form of free media, and at the same time, the automatic coloring task of hand-drawn sketches has become one of the research hotspots in the field of vision. Sketch coloring is a challenging task, mainly due to: a) lack of inherent color information; b) the result of coloring is prone to cross-color and overflow; c) poor controllability of the coloring process. Based on this, we design a unique deep network that combines self-attention mechanism to alleviate this problem in this paper. Specifically, we first use the encoding and decoding structure of U-net to capture the delicate details in the sketch; a unique self-attention module is then integrated to automatically capture key areas of the input sketch, thereby improving color distribution modeling and contextual awareness, enhancing the results of automatic coloring. Experiments on a large-scale anime sketch dataset show that the proposed method exhibits significant performance improvement, capturing delicate details in sketches while maintaining faithful color restoration. In addition, our model also supports guided coloring based on provided clues, which providing users more controllability.
- Open Access
- 18 July 2023
- Published by Elsevier in Graphical Models
- Vol. 129
Component-aware generative autoencoder for structure hybrid and shape completion
- 18 July 2023
- Published by Elsevier in Graphical Models
- Vol. 129
Open AccessComponent-aware generative autoencoder for structure hybrid and shape completion
- 27 October 2022
- Published by Springer Nature
- p. 668-680
TIR: A Two-Stage Insect Recognition Method for Convolutional Neural Network
- 27 October 2022
- Published by Springer Nature
- p. 668-680
TIR: A Two-Stage Insect Recognition Method for Convolutional Neural Network
- 27 October 2022
- Published by Springer Nature
- p. 130-145
General High-Pass Convolution: A Novel Convolutional Layer for Image Manipulation Detection
- 27 October 2022
- Published by Springer Nature
- p. 130-145
General High-Pass Convolution: A Novel Convolutional Layer for Image Manipulation Detection