1 citations · 1 across the 9 of their papers we have counts for
10 papers
Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
Zixiang Zhou, Zhentao Yu, Yifeng Ma +9
Subject-driven and multi-element video generation are central to controllable video synthesis, but existing methods still struggle to preserve identity consistency and model comple…
Goku: A Million-Scale Universal Dataset and Benchmark for Instruction-Based Video Editing
Sen Liang, Cong Wang, Zhentao Yu +8
Existing instruction-based video editing datasets commonly focus on single-task appearance editing, failing to meet the complex creative demands of real-world scenarios. To bridge…
HarmoView: Harmonizing Multi-View Constraints for Identity-Consistent Video Generation
Cong Wang, Zhentao Yu, Hongmei Wang +7
Current identity-consistent video generation methods struggle to preserve appearance fidelity under large viewpoint changes. While introducing multi-view reference input offers a n…
SpongeBob: Sync-Aware Harmonious Audio-Visual Generative Editing
Sen Liang, Cong Wang, Fengbin Guan +6
Visual and acoustic events in the physical world are inherently coupled, yet existing video editing methods typically adopt decoupled pipelines, lacking bidirectional modality inte…
Making Avatars Interact: Towards Text-Driven Human-Object Interaction for Controllable Talking Avatars
Youliang Zhang, Zhengguang Zhou, Zhentao Yu +11
Generating talking avatars is a fundamental task in video generation. Although existing methods can generate full-body talking avatars with simple human motion, extending this task…
Harmony: Harmonizing Audio and Video Generation through Cross-Task Synergy
Teng Hu, Zhentao Yu, Guozhen Zhang +6
The synthesis of synchronized audio-visual content is a key challenge in generative AI, with open-source models facing challenges in robust audio-video alignment. Our analysis reve…
PolyVivid: Vivid Multi-Subject Video Generation with Cross-Modal Interaction and Enhancement
Teng Hu, Zhentao Yu, Zhengguang Zhou +4
Despite recent advances in video generation, existing models still lack fine-grained controllability, especially for multi-subject customization with consistent identity and intera…
OmniV2V: Versatile Video Generation and Editing via Dynamic Content Manipulation
Sen Liang, Zhentao Yu, Zhengguang Zhou +8
The emergence of Diffusion Transformers (DiT) has brought significant advancements to video generation, especially in text-to-video and image-to-video tasks. Although video generat…
HunyuanCustom: A Multimodal-Driven Architecture for Customized Video Generation
Teng Hu, Zhentao Yu, Zhengguang Zhou +4
Customized video generation aims to produce videos featuring specific subjects under flexible user-defined conditions, yet existing methods often struggle with identity consistency…
HunyuanPortrait: Implicit Condition Control for Enhanced Portrait Animation
Zunnan Xu, Zhentao Yu, Zixiang Zhou +10
We introduce HunyuanPortrait, a diffusion-based condition control method that employs implicit representations for highly controllable and lifelike portrait animation. Given a sing…