Guangcong Zheng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Guangcong Zheng has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Xi Li
Zhejiang University, Hangzhou, China
5 shared publications Rui Jiang
College of Computer Science and Technology Zhejiang University
3 shared publications- Taiping Yao
YouTu Lab, Tencent, Shanghai, China
3 shared publications - Teng Li
College of Computer Science and Technology, Zhejiang University
3 shared publications Xi Li
College of Computer Science & Technology Zhejiang University
3 shared publications- Xinghe Fu
College of Computer Science and Technology, Zhejiang University, Hangzhou, China
2 shared publications
Latest Publications
- Open Access
Compact Attention: Exploiting Structured Spatio-Temporal Sparsity for Fast Video Generation
The computational demands of self-attention mechanisms pose a critical challenge for transformer-based video generation, particularly in synthesizing ultra-long sequences. Current approaches, such as factorized attention and fixed sparse patterns, fail to fully exploit the inherent spatio-temporal redundancies in video data. Through systematic analysis of video diffusion transformers (DiT), we uncover a key insight: Attention matrices exhibit structured, yet heterogeneous sparsity patterns, where specialized heads dynamically attend to distinct spatiotemporal regions (e.g., local pattern, cross-shaped pattern, or global pattern). Existing sparse attention methods either impose rigid constraints or introduce significant overhead, limiting their effectiveness. To address this, we propose Compact Attention, a hardware-aware acceleration framework featuring three innovations: 1) Adaptive tiling strategies that approximate diverse spatial interaction patterns via dynamic tile grouping, 2) Temporally varying windows that adjust sparsity levels based on frame proximity, and 3) An automated configuration search algorithm that optimizes sparse patterns while preserving critical attention pathways. Our method achieves 1.6~2.5x acceleration in attention computation on single-GPU setups while maintaining comparable visual quality with full-attention baselines. This work provides a principled approach to unlocking efficient long-form video generation through structured sparsity exploitation. Project Page: https://yo-ava.github.io/Compact-Attention.github.io/
Open AccessCompact Attention: Exploiting Structured Spatio-Temporal Sparsity for Fast Video Generation
The computational demands of self-attention mechanisms pose a critical challenge for transformer-based video generation, particularly in synthesizing ultra-long sequences. Current approaches, such as factorized attention and fixed sparse patterns, fail to fully exploit the inherent spatio-temporal redundancies in video data. Through systematic analysis of video diffusion transformers (DiT), we uncover a key insight: Attention matrices exhibit structured, yet heterogeneous sparsity patterns, where specialized heads dynamically attend to distinct spatiotemporal regions (e.g., local pattern, cross-shaped pattern, or global pattern). Existing sparse attention methods either impose rigid constraints or introduce significant overhead, limiting their effectiveness. To address this, we propose Compact Attention, a hardware-aware acceleration framework featuring three innovations: 1) Adaptive tiling strategies that approximate diverse spatial interaction patterns via dynamic tile grouping, 2) Temporally varying windows that adjust sparsity levels based on frame proximity, and 3) An automated configuration search algorithm that optimizes sparse patterns while preserving critical attention pathways. Our method achieves 1.6~2.5x acceleration in attention computation on single-GPU setups while maintaining comparable visual quality with full-attention baselines. This work provides a principled approach to unlocking efficient long-form video generation through structured sparsity exploitation. Project Page: https://yo-ava.github.io/Compact-Attention.github.io/
- Open Access
Dynamic-DINO: Fine-Grained Mixture of Experts Tuning for Real-time Open-Vocabulary Object Detection
The Mixture of Experts (MoE) architecture has excelled in Large Vision-Language Models (LVLMs), yet its potential in real-time open-vocabulary object detectors, which also leverage large-scale vision-language datasets but smaller models, remains unexplored. This work investigates this domain, revealing intriguing insights. In the shallow layers, experts tend to cooperate with diverse peers to expand the search space. While in the deeper layers, fixed collaborative structures emerge, where each expert maintains 2-3 fixed partners and distinct expert combinations are specialized in processing specific patterns. Concretely, we propose Dynamic-DINO, which extends Grounding DINO 1.5 Edge from a dense model to a dynamic inference framework via an efficient MoE-Tuning strategy. Additionally, we design a granularity decomposition mechanism to decompose the Feed-Forward Network (FFN) of base model into multiple smaller expert networks, expanding the subnet search space. To prevent performance degradation at the start of fine-tuning, we further propose a pre-trained weight allocation strategy for the experts, coupled with a specific router initialization. During inference, only the input-relevant experts are activated to form a compact subnet. Experiments show that, pretrained with merely 1.56M open-source data, Dynamic-DINO outperforms Grounding DINO 1.5 Edge, pretrained on the private Grounding20M dataset.
Open AccessDynamic-DINO: Fine-Grained Mixture of Experts Tuning for Real-time Open-Vocabulary Object Detection
The Mixture of Experts (MoE) architecture has excelled in Large Vision-Language Models (LVLMs), yet its potential in real-time open-vocabulary object detectors, which also leverage large-scale vision-language datasets but smaller models, remains unexplored. This work investigates this domain, revealing intriguing insights. In the shallow layers, experts tend to cooperate with diverse peers to expand the search space. While in the deeper layers, fixed collaborative structures emerge, where each expert maintains 2-3 fixed partners and distinct expert combinations are specialized in processing specific patterns. Concretely, we propose Dynamic-DINO, which extends Grounding DINO 1.5 Edge from a dense model to a dynamic inference framework via an efficient MoE-Tuning strategy. Additionally, we design a granularity decomposition mechanism to decompose the Feed-Forward Network (FFN) of base model into multiple smaller expert networks, expanding the subnet search space. To prevent performance degradation at the start of fine-tuning, we further propose a pre-trained weight allocation strategy for the experts, coupled with a specific router initialization. During inference, only the input-relevant experts are activated to form a compact subnet. Experiments show that, pretrained with merely 1.56M open-source data, Dynamic-DINO outperforms Grounding DINO 1.5 Edge, pretrained on the private Grounding20M dataset.
- Top Cited PapersOpen Access
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (8) , 8469-8477
CustomCrafter: Customized Video Generation with Preserving Motion and Concept Composition Abilities
Customized video generation aims to generate high-quality videos guided by text prompts and subject's reference images. However, since it is only trained on static images, the fine-tuning process of subject learning disrupts abilities of video diffusion models (VDMs) to combine concepts and generate motions. To restore these abilities, some methods use additional video similar to the prompt to fine-tune or guide the model. This requires frequent changes of guiding videos and even re-tuning of the model when generating different motions, which is very inconvenient for users. In this paper, we propose CustomCrafter, a novel framework that preserves the model's motion generation and conceptual combination abilities without additional video and fine-tuning to recovery. For preserving conceptual combination ability, we design a plug-and-play module to update few parameters in VDMs, enhancing the model's ability to capture the appearance details and the ability of concept combinations for new subjects. For motion generation, we observed that VDMs tend to restore the motion of video in the early stage of denoising, while focusing on the recovery of subject details in the later stage. Therefore, we propose Dynamic Weighted Video Sampling Strategy. Using the pluggability of our subject learning modules, we reduce the impact of this module on motion generation in the early stage of denoising, preserving the ability to generate motion of VDMs. In the later stage of denoising, we restore this module to repair the appearance details of the specified subject, thereby ensuring the fidelity of the subject's appearance. Experimental results show that our method has a significant improvement compared to previous methods.
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (8) , 8469-8477
Top Cited PapersOpen AccessCustomCrafter: Customized Video Generation with Preserving Motion and Concept Composition Abilities
Customized video generation aims to generate high-quality videos guided by text prompts and subject's reference images. However, since it is only trained on static images, the fine-tuning process of subject learning disrupts abilities of video diffusion models (VDMs) to combine concepts and generate motions. To restore these abilities, some methods use additional video similar to the prompt to fine-tune or guide the model. This requires frequent changes of guiding videos and even re-tuning of the model when generating different motions, which is very inconvenient for users. In this paper, we propose CustomCrafter, a novel framework that preserves the model's motion generation and conceptual combination abilities without additional video and fine-tuning to recovery. For preserving conceptual combination ability, we design a plug-and-play module to update few parameters in VDMs, enhancing the model's ability to capture the appearance details and the ability of concept combinations for new subjects. For motion generation, we observed that VDMs tend to restore the motion of video in the early stage of denoising, while focusing on the recovery of subject details in the later stage. Therefore, we propose Dynamic Weighted Video Sampling Strategy. Using the pluggability of our subject learning modules, we reduce the impact of this module on motion generation in the early stage of denoising, preserving the ability to generate motion of VDMs. In the later stage of denoising, we restore this module to repair the appearance details of the specified subject, thereby ensuring the fidelity of the subject's appearance. Experimental results show that our method has a significant improvement compared to previous methods.
- Open Access
RealCam-Vid: High-resolution Video Dataset with Dynamic Scenes and Metric-scale Camera Movements
Recent advances in camera-controllable video generation have been constrained by the reliance on static-scene datasets with relative-scale camera annotations, such as RealEstate10K. While these datasets enable basic viewpoint control, they fail to capture dynamic scene interactions and lack metric-scale geometric consistency-critical for synthesizing realistic object motions and precise camera trajectories in complex environments. To bridge this gap, we introduce the first fully open-source, high-resolution dynamic-scene dataset with metric-scale camera annotations in https://github.com/ZGCTroy/RealCam-Vid.
Open AccessRealCam-Vid: High-resolution Video Dataset with Dynamic Scenes and Metric-scale Camera Movements
Recent advances in camera-controllable video generation have been constrained by the reliance on static-scene datasets with relative-scale camera annotations, such as RealEstate10K. While these datasets enable basic viewpoint control, they fail to capture dynamic scene interactions and lack metric-scale geometric consistency-critical for synthesizing realistic object motions and precise camera trajectories in complex environments. To bridge this gap, we introduce the first fully open-source, high-resolution dynamic-scene dataset with metric-scale camera annotations in https://github.com/ZGCTroy/RealCam-Vid.
- Open Access
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (4) , 4048-4056
Energy-Guided Optimization for Personalized Image Editing with Pretrained Text-to-Image Diffusion Models
The rapid advancement of pretrained text-driven diffusion models has significantly enriched applications in image generation and editing. However, as the demand for personalized content editing increases, new challenges emerge especially when dealing with arbitrary objects and complex scenes. Existing methods usually mistakes mask as the object shape prior, which struggle to achieve a seamless integration result. The mostly used inversion noise initialization also hinders the identity consistency towards the target object. To address these challenges, we propose a novel training-free framework that formulates personalized content editing as the optimization of edited images in the latent space, using diffusion models as the energy function guidance conditioned by reference text-image pairs. A coarse-to-fine strategy is proposed that employs text energy guidance at the early stage to achieve a natural transition toward the target class and uses point-to-point feature-level image energy guidance to perform fine-grained appearance alignment with the target object. Additionally, we introduce the latent space content composition to enhance overall identity consistency with the target. Extensive experiments demonstrate that our method excels in object replacement even with a large domain gap, highlighting its potential for high-quality, personalized image editing.
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (4) , 4048-4056
Open AccessEnergy-Guided Optimization for Personalized Image Editing with Pretrained Text-to-Image Diffusion Models
The rapid advancement of pretrained text-driven diffusion models has significantly enriched applications in image generation and editing. However, as the demand for personalized content editing increases, new challenges emerge especially when dealing with arbitrary objects and complex scenes. Existing methods usually mistakes mask as the object shape prior, which struggle to achieve a seamless integration result. The mostly used inversion noise initialization also hinders the identity consistency towards the target object. To address these challenges, we propose a novel training-free framework that formulates personalized content editing as the optimization of edited images in the latent space, using diffusion models as the energy function guidance conditioned by reference text-image pairs. A coarse-to-fine strategy is proposed that employs text energy guidance at the early stage to achieve a natural transition toward the target class and uses point-to-point feature-level image energy guidance to perform fine-grained appearance alignment with the target object. Additionally, we introduce the latent space content composition to enhance overall identity consistency with the target. Extensive experiments demonstrate that our method excels in object replacement even with a large domain gap, highlighting its potential for high-quality, personalized image editing.
- Open Access
Energy-Guided Optimization for Personalized Image Editing with Pretrained Text-to-Image Diffusion Models
The rapid advancement of pretrained text-driven diffusion models has significantly enriched applications in image generation and editing. However, as the demand for personalized content editing increases, new challenges emerge especially when dealing with arbitrary objects and complex scenes. Existing methods usually mistakes mask as the object shape prior, which struggle to achieve a seamless integration result. The mostly used inversion noise initialization also hinders the identity consistency towards the target object. To address these challenges, we propose a novel training-free framework that formulates personalized content editing as the optimization of edited images in the latent space, using diffusion models as the energy function guidance conditioned by reference text-image pairs. A coarse-to-fine strategy is proposed that employs text energy guidance at the early stage to achieve a natural transition toward the target class and uses point-to-point feature-level image energy guidance to perform fine-grained appearance alignment with the target object. Additionally, we introduce the latent space content composition to enhance overall identity consistency with the target. Extensive experiments demonstrate that our method excels in object replacement even with a large domain gap, highlighting its potential for high-quality, personalized image editing.
Open AccessEnergy-Guided Optimization for Personalized Image Editing with Pretrained Text-to-Image Diffusion Models
The rapid advancement of pretrained text-driven diffusion models has significantly enriched applications in image generation and editing. However, as the demand for personalized content editing increases, new challenges emerge especially when dealing with arbitrary objects and complex scenes. Existing methods usually mistakes mask as the object shape prior, which struggle to achieve a seamless integration result. The mostly used inversion noise initialization also hinders the identity consistency towards the target object. To address these challenges, we propose a novel training-free framework that formulates personalized content editing as the optimization of edited images in the latent space, using diffusion models as the energy function guidance conditioned by reference text-image pairs. A coarse-to-fine strategy is proposed that employs text energy guidance at the early stage to achieve a natural transition toward the target class and uses point-to-point feature-level image energy guidance to perform fine-grained appearance alignment with the target object. Additionally, we introduce the latent space content composition to enhance overall identity consistency with the target. Extensive experiments demonstrate that our method excels in object replacement even with a large domain gap, highlighting its potential for high-quality, personalized image editing.
- 1 May 2024
- Published by Springer Nature in Journal of Computer Science and Technology
- Vol. 39 (3) , 509-541
A Survey of Multimodal Controllable Diffusion Models
- 1 May 2024
- Published by Springer Nature in Journal of Computer Science and Technology
- Vol. 39 (3) , 509-541
A Survey of Multimodal Controllable Diffusion Models
- 23 October 2022
- Published by Springer Nature
- p. 754-769
Entropy-Driven Sampling and Training Scheme for Conditional Diffusion Generation
- 23 October 2022
- Published by Springer Nature
- p. 754-769
Entropy-Driven Sampling and Training Scheme for Conditional Diffusion Generation