Bo Li
Research Subjects
Bo Li has a total of 119 co-authors. Here are the top co-authors by number of shared publications:
- Peng-Tao Jiang
Vivo
10 shared publications - Hao Zhang
Vivo Mobile Communication Co., Ltd, Shanghai, China
6 shared publications Shouhong Ding
Youtu Laboratory, Tencent, Shanghai, China
6 shared publications- Tianyi Zheng
Shanghai Jiao Tong University
6 shared publications Lv Tang
Vivo Mobile Communication Co., Ltd, Shanghai, China; Department of Electrical and Computer Engineering, University of Alberta, Edmonton, Canada
5 shared publications- Lv Tang
School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing, China
5 shared publications
Latest Publications
- 25 September 2026
- Published by Springer Nature
- p. 399-416
Anchor Forcing: Anchor Memory and Tri-Region RoPE for Interactive Streaming Video Diffusion
- 25 September 2026
- Published by Springer Nature
- p. 399-416
Anchor Forcing: Anchor Memory and Tri-Region RoPE for Interactive Streaming Video Diffusion
- 15 September 2026
- Published by Springer Nature
- p. 185-202
GR-GRPO: Graph-Diffused Credit for Autoregressive Image RL Alignment
- 15 September 2026
- Published by Springer Nature
- p. 185-202
GR-GRPO: Graph-Diffused Credit for Autoregressive Image RL Alignment
- 12 September 2026
- Published by Springer Nature
- p. 601-618
SyncCache: Exploiting Asymmetric Dynamics for Fast Audio-Driven Portrait Animation
- 12 September 2026
- Published by Springer Nature
- p. 601-618
SyncCache: Exploiting Asymmetric Dynamics for Fast Audio-Driven Portrait Animation
- 11 September 2026
- Published by Springer Nature
- p. 38-55
MagicMakeup: A Region-Controllable Diffusion Transformer for High-Fidelity Makeup-Transfer
- 11 September 2026
- Published by Springer Nature
- p. 38-55
MagicMakeup: A Region-Controllable Diffusion Transformer for High-Fidelity Makeup-Transfer
- 10 September 2026
- Published by Springer Nature
- p. 278-298
EditHF-1M: A Million-Scale Rich Human Preference Feedback for Image Editing
- 10 September 2026
- Published by Springer Nature
- p. 278-298
EditHF-1M: A Million-Scale Rich Human Preference Feedback for Image Editing
- 10 September 2026
- Published by Springer Nature
- p. 520-538
JSON: Jigsaw Self-play Optimization for Normalizing Flows
- 10 September 2026
- Published by Springer Nature
- p. 520-538
JSON: Jigsaw Self-play Optimization for Normalizing Flows
- Top Cited Papers
- 6 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. PP  (10518215) , 1
DepthMaster: Taming Diffusion Models for Monocular Depth Estimation
Monocular depth estimation within the denoising diffusion paradigm demonstrates impressive generalization ability but suffers from low inference speed. Recent methods adopt a single-step deterministic paradigm to improve inference efficiency. However, uncritically applying the generative features from diffusion models to the perceptual depth estimation task leads to suboptimal results, due to the differences between image generation and dense prediction. In this work, we propose Depth-Master, a single-step diffusion model designed to tame diffusion models for enhanced performance and visual quality. First, to mitigate overfitting to texture details introduced by generative features, we propose a Feature Alignment module, which incorporates high-quality semantic features to enhance the denoising network’s representation capability. Second, to address the lack of fine-grained details in the single-step deterministic framework, we propose a Fourier Enhancement module to adaptively balance low-frequency structure and high-frequency details. We adopt a two-stage training strategy to fully leverage the potential of the two modules. In the first stage, we focus on learning the global scene structure with the Feature Alignment module, while in the second stage, we exploit the Fourier Enhancement module to improve the visual quality. Through these efforts, our model achieves state-of-the-art performance in terms of generalization and detail preservation, outperforming other diffusion-based methods across various datasets. Our project page can be found at https://indu1ge.github.io/DepthMaster page.
- 6 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. PP  (10518215) , 1
Top Cited PapersDepthMaster: Taming Diffusion Models for Monocular Depth Estimation
Monocular depth estimation within the denoising diffusion paradigm demonstrates impressive generalization ability but suffers from low inference speed. Recent methods adopt a single-step deterministic paradigm to improve inference efficiency. However, uncritically applying the generative features from diffusion models to the perceptual depth estimation task leads to suboptimal results, due to the differences between image generation and dense prediction. In this work, we propose Depth-Master, a single-step diffusion model designed to tame diffusion models for enhanced performance and visual quality. First, to mitigate overfitting to texture details introduced by generative features, we propose a Feature Alignment module, which incorporates high-quality semantic features to enhance the denoising network’s representation capability. Second, to address the lack of fine-grained details in the single-step deterministic framework, we propose a Fourier Enhancement module to adaptively balance low-frequency structure and high-frequency details. We adopt a two-stage training strategy to fully leverage the potential of the two modules. In the first stage, we focus on learning the global scene structure with the Feature Alignment module, while in the second stage, we exploit the Fourier Enhancement module to improve the visual quality. Through these efforts, our model achieves state-of-the-art performance in terms of generalization and detail preservation, outperforming other diffusion-based methods across various datasets. Our project page can be found at https://indu1ge.github.io/DepthMaster page.
- Open Access
AnimeAgent: Is the Multi-Agent via Image-to-Video models a Good Disney Storytelling Artist?
Custom Storyboard Generation (CSG) aims to produce high-quality, multi-character consistent storytelling. Current approaches based on static diffusion models, whether used in a one-shot manner or within multi-agent frameworks, face three key limitations: (1) Static models lack dynamic expressiveness and often resort to "copy-paste" pattern. (2) One-shot inference cannot iteratively correct missing attributes or poor prompt adherence. (3) Multi-agents rely on non-robust evaluators, ill-suited for assessing stylized, non-realistic animation. To address these, we propose AnimeAgent, the first Image-to-Video (I2V)-based multi-agent framework for CSG. Inspired by Disney's "Combination of Straight Ahead and Pose to Pose" workflow, AnimeAgent leverages I2V's implicit motion prior to enhance consistency and expressiveness, while a mixed subjective-objective reviewer enables reliable iterative refinement. We also collect a human-annotated CSG benchmark with ground-truth. Experiments show AnimeAgent achieves SOTA performance in consistency, prompt fidelity, and stylization.
Open AccessAnimeAgent: Is the Multi-Agent via Image-to-Video models a Good Disney Storytelling Artist?
Custom Storyboard Generation (CSG) aims to produce high-quality, multi-character consistent storytelling. Current approaches based on static diffusion models, whether used in a one-shot manner or within multi-agent frameworks, face three key limitations: (1) Static models lack dynamic expressiveness and often resort to "copy-paste" pattern. (2) One-shot inference cannot iteratively correct missing attributes or poor prompt adherence. (3) Multi-agents rely on non-robust evaluators, ill-suited for assessing stylized, non-realistic animation. To address these, we propose AnimeAgent, the first Image-to-Video (I2V)-based multi-agent framework for CSG. Inspired by Disney's "Combination of Straight Ahead and Pose to Pose" workflow, AnimeAgent leverages I2V's implicit motion prior to enhance consistency and expressiveness, while a mixed subjective-objective reviewer enables reliable iterative refinement. We also collect a human-annotated CSG benchmark with ground-truth. Experiments show AnimeAgent achieves SOTA performance in consistency, prompt fidelity, and stylization.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 12864-12875
Improving Consistency in Diffusion Models for Image Super-Resolution
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 12864-12875
Improving Consistency in Diffusion Models for Image Super-Resolution
- 26 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36  (4) , 5329-5343
SEMat: Semantic Enhanced Natural Image Interactive Matting
Recent approaches attempt to adapt powerful interactive segmentation models, such as SAM, to interactive matting and fine-tune the models based on synthetic matting datasets. However, models trained on synthetic data fail to generalize to complex and occlusion scenes. We address this challenge by proposing a new matting dataset based on the COCO dataset, namely COCO-Matting. It selects real-world complex images from COCO and converts semantic segmentation masks to matting labels. The built COCO-Matting comprises an extensive collection of 36,980 human instance-level alpha mattes in complex natural scenarios. Furthermore, existing SAM-based matting methods extract intermediate features and masks from a frozen SAM and only train a lightweight matting decoder by end-to-end matting losses, which do not fully exploit the potential of the pre-trained SAM. Thus, we propose SEMat which revamps the network architecture and training objectives. For network architecture, the proposed feature-aligned transformer learns to extract fine-grained edge and transparency features. The proposed matte-aligned decoder aims to segment matting-specific objects and convert coarse masks into high-precision mattes. For training objectives, the proposed regularization and trimap loss aim to retain the prior from the pre-trained model and push the matting logits extracted from the mask decoder to contain trimap-based semantic information. Extensive experiments across seven diverse datasets demonstrate the superior performance of our method, proving its efficacy in interactive natural image matting. Code is available at https://github.com/XiaRho/SEMat.
- 26 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36  (4) , 5329-5343
SEMat: Semantic Enhanced Natural Image Interactive Matting
Recent approaches attempt to adapt powerful interactive segmentation models, such as SAM, to interactive matting and fine-tune the models based on synthetic matting datasets. However, models trained on synthetic data fail to generalize to complex and occlusion scenes. We address this challenge by proposing a new matting dataset based on the COCO dataset, namely COCO-Matting. It selects real-world complex images from COCO and converts semantic segmentation masks to matting labels. The built COCO-Matting comprises an extensive collection of 36,980 human instance-level alpha mattes in complex natural scenarios. Furthermore, existing SAM-based matting methods extract intermediate features and masks from a frozen SAM and only train a lightweight matting decoder by end-to-end matting losses, which do not fully exploit the potential of the pre-trained SAM. Thus, we propose SEMat which revamps the network architecture and training objectives. For network architecture, the proposed feature-aligned transformer learns to extract fine-grained edge and transparency features. The proposed matte-aligned decoder aims to segment matting-specific objects and convert coarse masks into high-precision mattes. For training objectives, the proposed regularization and trimap loss aim to retain the prior from the pre-trained model and push the matting logits extracted from the mask decoder to contain trimap-based semantic information. Extensive experiments across seven diverse datasets demonstrate the superior performance of our method, proving its efficacy in interactive natural image matting. Code is available at https://github.com/XiaRho/SEMat.