Zhoujie Fu

Yunze Liu, Yun Liu, Che Jiang et al.

We present HOI4D, a large-scale 4D egocentric dataset with rich annotations, to catalyze the research of category-level human-object interaction. HOI4D consists of 2.4M RGB-D egocentric video frames over 4000 sequences collected by 4 participants interacting with 800 different object instances from 16 categories over 610 different indoor rooms. Frame-wise annotations for panoptic segmentation, motion segmentation, 3D hand pose, category-level object pose and hand action have also been provided, together with reconstructed object meshes and scene point clouds. With HOI4D, we establish three benchmarking tasks to promote category-level HOI from 4D visual signals including semantic segmentation of 4D dynamic point cloud sequences, category-level object pose tracking, and egocentric action segmentation with diverse interaction targets. In-depth analysis shows HOI4D poses great challenges to existing methods and produces great research opportunities.

Fan Wu, Cheng Chen, Zhoujie Fu et al.

Recent diffusion-based text-to-image customization methods have achieved significant success in understanding concrete concepts to control generation processes, such as styles and shapes. However, few efforts dive into the realistic yet challenging customization of physical concepts. The core limitation of current methods arises from the absence of explicitly introducing physical knowledge during training. Even when physics-related words appear in the input text prompts, our experiments consistently demonstrate that these methods fail to accurately reflect the corresponding physical properties in the generated results. In this paper, we propose PhyCustom, a fine-tuning framework comprising two novel regularization losses to activate diffusion model to perform physical customization. Specifically, the proposed isometric loss aims at activating diffusion models to learn physical concepts while decouple loss helps to eliminate the mixture learning of independent concepts. Experiments are conducted on a diverse dataset and our benchmark results demonstrate that PhyCustom outperforms previous state-of-the-art and popular methods in terms of physical customization quantitatively and qualitatively.

Junyi Chen, Tong He, Zhoujie Fu et al.

We present VINO, a unified visual generator that performs image and video generation and editing within a single framework. Instead of relying on task-specific models or independent modules for each modality, VINO uses a shared diffusion backbone that conditions on text, images and videos, enabling a broad range of visual creation and editing tasks under one model. Specifically, VINO couples a vision-language model (VLM) with a Multimodal Diffusion Transformer (MMDiT), where multimodal inputs are encoded as interleaved conditioning tokens, and then used to guide the diffusion process. This design supports multi-reference grounding, long-form instruction following, and coherent identity preservation across static and dynamic content, while avoiding modality-specific architectural components. To train such a unified system, we introduce a multi-stage training pipeline that progressively expands a video generation base model into a unified, multi-task generator capable of both image and video input and output. Across diverse generation and editing benchmarks, VINO demonstrates strong visual quality, faithful instruction following, improved reference and attribute preservation, and more controllable multi-identity edits. Our results highlight a practical path toward scalable unified visual generation, and the promise of interleaved, in-context computation as a foundation for general-purpose visual creation.

Tianyu Sun, Zhoujie Fu, Bang Zhang et al.

The demand for realistic and versatile character animation has surged, driven by its wide-ranging applications in various domains. However, the animation generation algorithms modeling human pose with 2D or 3D structures all face various problems, including low-quality output content and training data deficiency, preventing the related algorithms from generating high-quality animation videos. Therefore, we introduce MVAnimate, a novel framework that synthesizes both 2D and 3D information of dynamic figures based on multi-view prior information, to enhance the generated video quality. Our approach leverages multi-view prior information to produce temporally consistent and spatially coherent animation outputs, demonstrating improvements over existing animation methods. Our MVAnimate also optimizes the multi-view videos of the target character, enhancing the video quality from different views. Experimental results on diverse datasets highlight the robustness of our method in handling various motion patterns and appearances.

Junyi Chen, Haoyi Zhu, Xianglong He et al.

World models serve as essential building blocks toward Artificial General Intelligence (AGI), enabling intelligent agents to predict future states and plan actions by simulating complex physical interactions. However, existing interactive models primarily predict visual observations, thereby neglecting crucial hidden states like geometric structures and spatial coherence. This leads to rapid error accumulation and temporal inconsistency. To address these limitations, we introduce DeepVerse, a novel 4D interactive world model explicitly incorporating geometric predictions from previous timesteps into current predictions conditioned on actions. Experiments demonstrate that by incorporating explicit geometric constraints, DeepVerse captures richer spatio-temporal relationships and underlying physical dynamics. This capability significantly reduces drift and enhances temporal consistency, enabling the model to reliably generate extended future sequences and achieve substantial improvements in prediction accuracy, visual realism, and scene rationality. Furthermore, our method provides an effective solution for geometry-aware memory retrieval, effectively preserving long-term spatial consistency. We validate the effectiveness of DeepVerse across diverse scenarios, establishing its capacity for high-fidelity, long-horizon predictions grounded in geometry-aware dynamics.

Zhoujie Fu, Xianfang Zeng, Jinghong Lan et al.

Pre-trained video models learn powerful priors for generating high-quality, temporally coherent content. While these models excel at temporal coherence, their dynamics are often constrained by the continuous nature of their training data. We hypothesize that by injecting the rich and unconstrained content diversity from image data into this coherent temporal framework, we can generate image sets that feature both natural transitions and a far more expansive dynamic range. To this end, we introduce iMontage, a unified framework designed to repurpose a powerful video model into an all-in-one image generator. The framework consumes and produces variable-length image sets, unifying a wide array of image generation and editing tasks. To achieve this, we propose an elegant and minimally invasive adaptation strategy, complemented by a tailored data curation process and training paradigm. This approach allows the model to acquire broad image manipulation capabilities without corrupting its invaluable original motion priors. iMontage excels across several mainstream many-in-many-out tasks, not only maintaining strong cross-image contextual consistency but also generating scenes with extraordinary dynamics that surpass conventional scopes. Find our homepage at: https://kr1sjfu.github.io/iMontage-web/.

Xinyao Liao, Xianfang Zeng, Ziye Song et al.

Despite the rapid progress of instruction-based image editing, its extension to video remains underexplored, primarily due to the prohibitive cost and complexity of constructing large-scale paired video editing datasets. To address this challenge, we introduce a low-cost pretraining strategy for instruction-based video editing that leverages in-context learning from unpaired video clips. We show that pretraining a foundation video generation model with this strategy endows it with general editing capabilities, such as adding, replacing, or deleting operations, according to input editing instructions. The pretrained model can then be efficiently refined with a small amount of high-quality paired editing data. Built upon HunyuanVideoT2V, our framework first pretrains on approximately 1M real video clips to learn basic editing concepts, and subsequently fine-tunes on fewer than 150k curated editing pairs to extend more editing tasks and improve the editing quality. Comparative experiments show that our method surpasses existing instruction-based video editing approaches in both instruction alignment and visual fidelity, achieving a 12\% improvement in editing instruction following and a 15\% improvement in editing quality.

Yang Zhou, Yifan Wang, Jianjun Zhou et al.

The field of 4D world modeling - aiming to jointly capture spatial geometry and temporal dynamics - has witnessed remarkable progress in recent years, driven by advances in large-scale generative models and multimodal learning. However, the development of truly general 4D world models remains fundamentally constrained by the availability of high-quality data. Existing datasets and benchmarks often lack the dynamic complexity, multi-domain diversity, and spatial-temporal annotations required to support key tasks such as 4D geometric reconstruction, future prediction, and camera-control video generation. To address this gap, we introduce OmniWorld, a large-scale, multi-domain, multi-modal dataset specifically designed for 4D world modeling. OmniWorld consists of a newly collected OmniWorld-Game dataset and several curated public datasets spanning diverse domains. Compared with existing synthetic datasets, OmniWorld-Game provides richer modality coverage, larger scale, and more realistic dynamic interactions. Based on this dataset, we establish a challenging benchmark that exposes the limitations of current state-of-the-art (SOTA) approaches in modeling complex 4D environments. Moreover, fine-tuning existing SOTA methods on OmniWorld leads to significant performance gains across 4D reconstruction and video generation tasks, strongly validating OmniWorld as a powerful resource for training and evaluation. We envision OmniWorld as a catalyst for accelerating the development of general-purpose 4D world models, ultimately advancing machines' holistic understanding of the physical world.

Cheng Chen, Xiaofeng Yang, Fan Yang et al.

Recent works on text-to-3d generation show that using only 2D diffusion supervision for 3D generation tends to produce results with inconsistent appearances (e.g., faces on the back view) and inaccurate shapes (e.g., animals with extra legs). Existing methods mainly address this issue by retraining diffusion models with images rendered from 3D data to ensure multi-view consistency while struggling to balance 2D generation quality with 3D consistency. In this paper, we present a new framework Sculpt3D that equips the current pipeline with explicit injection of 3D priors from retrieved reference objects without re-training the 2D diffusion model. Specifically, we demonstrate that high-quality and diverse 3D geometry can be guaranteed by keypoints supervision through a sparse ray sampling approach. Moreover, to ensure accurate appearances of different views, we further modulate the output of the 2D diffusion model to the correct patterns of the template views without altering the generated object's style. These two decoupled designs effectively harness 3D information from reference objects to generate 3D objects while preserving the generation quality of the 2D diffusion model. Extensive experiments show our method can largely improve the multi-view consistency while retaining fidelity and diversity. Our project page is available at: https://stellarcheng.github.io/Sculpt3D/.

Youtan Yin, Zhoujie Fu, Fan Yang et al.

The emergence of Neural Radiance Fields (NeRF) for novel view synthesis has increased interest in 3D scene editing. An essential task in editing is removing objects from a scene while ensuring visual reasonability and multiview consistency. However, current methods face challenges such as time-consuming object labeling, limited capability to remove specific targets, and compromised rendering quality after removal. This paper proposes a novel object-removing pipeline, named OR-NeRF, that can remove objects from 3D scenes with user-given points or text prompts on a single view, achieving better performance in less time than previous works. Our method spreads user annotations to all views through 3D geometry and sparse correspondence, ensuring 3D consistency with less processing burden. Then recent 2D segmentation model Segment-Anything (SAM) is applied to predict masks, and a 2D inpainting model is used to generate color supervision. Finally, our algorithm applies depth supervision and perceptual loss to maintain consistency in geometry and appearance after object removal. Experimental results demonstrate that our method achieves better editing quality with less time than previous works, considering both quality and quantity.