arXiv subjects

Zheng-Jun Zha

Publications and source records attributed to Zheng-Jun Zha.

At least 73 records · Page 4Linked to original sources

arXiv2025

VideoGen-Eval: Agent-based System for Video Generation Evaluation

The rapid advancement of video generation has rendered existing evaluation systems inadequate for assessing state-of-the-art models, primarily due to simple prompts that cannot showcase the model's capabilities, fixed evaluation operators struggling with Out-of-Distribution (OOD) cases, and misalignment between computed metrics and human preferences. To bridge the gap, we propose VideoGen-Eval, an agent evaluation system that integrates LLM-based content structuring, MLLM-based content judgment, and patch tools designed for temporal-dense dimensions, to achieve a dynamic, flexible, and expandable video generation evaluation. Additionally, we introduce a video generation benchmark to evaluate existing cutting-edge models and verify the effectiveness of our evaluation system. It comprises 700 structured, content-rich prompts (both T2V and I2V) and over 12,000 videos generated by 20+ models, among them, 8 cutting-edge models are selected as quantitative evaluation for the agent and human. Extensive experiments validate that our proposed agent-based evaluation system demonstrates strong alignment with human preferences and reliably completes the evaluation, as well as the diversity and richness of the benchmark.

cs.CV↗

arXiv2025

NTIRE 2025 Challenge on Event-Based Image Deblurring: Methods and Results

This paper presents an overview of NTIRE 2025 the First Challenge on Event-Based Image Deblurring, detailing the proposed methodologies and corresponding results. The primary goal of the challenge is to design an event-based method that achieves high-quality image deblurring, with performance quantitatively assessed using Peak Signal-to-Noise Ratio (PSNR). Notably, there are no restrictions on computational complexity or model size. The task focuses on leveraging both events and images as inputs for single-image deblurring. A total of 199 participants registered, among whom 15 teams successfully submitted valid results, offering valuable insights into the current state of event-based image deblurring. We anticipate that this challenge will drive further advancements in event-based vision research.

cs.CV↗

arXiv2025

SIGMAN:Scaling 3D Human Gaussian Generation with Millions of Assets

3D human digitization has long been a highly pursued yet challenging task. Existing methods aim to generate high-quality 3D digital humans from single or multiple views, but remain primarily constrained by current paradigms and the scarcity of 3D human assets. Specifically, recent approaches fall into several paradigms: optimization-based and feed-forward (both single-view regression and multi-view generation with reconstruction). However, they are limited by slow speed, low quality, cascade reasoning, and ambiguity in mapping low-dimensional planes to high-dimensional space due to occlusion and invisibility, respectively. Furthermore, existing 3D human assets remain small-scale, insufficient for large-scale training. To address these challenges, we propose a latent space generation paradigm for 3D human digitization, which involves compressing multi-view images into Gaussians via a UV-structured VAE, along with DiT-based conditional generation, we transform the ill-posed low-to-high-dimensional mapping problem into a learnable distribution shift, which also supports end-to-end inference. In addition, we employ the multi-view optimization approach combined with synthetic data to construct the HGS-1M dataset, which contains $1$ million 3D Gaussian assets to support the large-scale training. Experimental results demonstrate that our paradigm, powered by large-scale training, produces high-quality 3D human Gaussians with intricate textures, facial details, and loose clothing deformation.

cs.CV↗

arXiv2025

EventMamba: Enhancing Spatio-Temporal Locality with State Space Models for Event-Based Video Reconstruction

Leveraging its robust linear global modeling capability, Mamba has notably excelled in computer vision. Despite its success, existing Mamba-based vision models have overlooked the nuances of event-driven tasks, especially in video reconstruction. Event-based video reconstruction (EBVR) demands spatial translation invariance and close attention to local event relationships in the spatio-temporal domain. Unfortunately, conventional Mamba algorithms apply static window partitions and standard reshape scanning methods, leading to significant losses in local connectivity. To overcome these limitations, we introduce EventMamba--a specialized model designed for EBVR tasks. EventMamba innovates by incorporating random window offset (RWO) in the spatial domain, moving away from the restrictive fixed partitioning. Additionally, it features a new consistent traversal serialization approach in the spatio-temporal domain, which maintains the proximity of adjacent events both spatially and temporally. These enhancements enable EventMamba to retain Mamba's robust modeling capabilities while significantly preserving the spatio-temporal locality of event data. Comprehensive testing on multiple datasets shows that EventMamba markedly enhances video reconstruction, drastically improving computation speed while delivering superior visual quality compared to Transformer-based methods.

cs.CV↗

arXiv2025

GREAT: Geometry-Intention Collaborative Inference for Open-Vocabulary 3D Object Affordance Grounding

Open-Vocabulary 3D object affordance grounding aims to anticipate ``action possibilities'' regions on 3D objects with arbitrary instructions, which is crucial for robots to generically perceive real scenarios and respond to operational changes. Existing methods focus on combining images or languages that depict interactions with 3D geometries to introduce external interaction priors. However, they are still vulnerable to a limited semantic space by failing to leverage implied invariant geometries and potential interaction intentions. Normally, humans address complex tasks through multi-step reasoning and respond to diverse situations by leveraging associative and analogical thinking. In light of this, we propose GREAT (GeometRy-intEntion collAboraTive inference) for Open-Vocabulary 3D Object Affordance Grounding, a novel framework that mines the object invariant geometry attributes and performs analogically reason in potential interaction scenarios to form affordance knowledge, fully combining the knowledge with both geometries and visual contents to ground 3D object affordance. Besides, we introduce the Point Image Affordance Dataset v2 (PIADv2), the largest 3D object affordance dataset at present to support the task. Extensive experiments demonstrate the effectiveness and superiority of GREAT. The code and dataset are available at https://yawen-shao.github.io/GREAT/.

cs.CV↗

arXiv2025

Benchmarking Large Vision-Language Models via Directed Scene Graph for Comprehensive Image Captioning

Generating detailed captions comprehending text-rich visual content in images has received growing attention for Large Vision-Language Models (LVLMs). However, few studies have developed benchmarks specifically tailored for detailed captions to measure their accuracy and comprehensiveness. In this paper, we introduce a detailed caption benchmark, termed as CompreCap, to evaluate the visual context from a directed scene graph view. Concretely, we first manually segment the image into semantically meaningful regions (i.e., semantic segmentation mask) according to common-object vocabulary, while also distinguishing attributes of objects within all those regions. Then directional relation labels of these objects are annotated to compose a directed scene graph that can well encode rich compositional information of the image. Based on our directed scene graph, we develop a pipeline to assess the generated detailed captions from LVLMs on multiple levels, including the object-level coverage, the accuracy of attribute descriptions, the score of key relationships, etc. Experimental results on the CompreCap dataset confirm that our evaluation method aligns closely with human evaluation scores across LVLMs.

cs.CV↗

arXiv2025

Collaborative Temporal Consistency Learning for Point-supervised Natural Language Video Localization

Natural language video localization (NLVL) is a crucial task in video understanding that aims to localize the target moment in videos specified by a given language description. Recently, a point-supervised paradigm has been presented to address this task, requiring only a single annotated frame within the target moment rather than complete temporal boundaries. Compared with the fully-supervised paradigm, it offers a balance between localization accuracy and annotation cost. However, due to the absence of complete annotation, it is challenging to align the video content with language descriptions, consequently hindering accurate moment prediction. To address this problem, we propose a new COllaborative Temporal consistEncy Learning (COTEL) framework that leverages the synergy between saliency detection and moment localization to strengthen the video-language alignment. Specifically, we first design a frame- and a segment-level Temporal Consistency Learning (TCL) module that models semantic alignment across frame saliencies and sentence-moment pairs. Then, we design a cross-consistency guidance scheme, including a Frame-level Consistency Guidance (FCG) and a Segment-level Consistency Guidance (SCG), that enables the two temporal consistency learning paths to reinforce each other mutually. Further, we introduce a Hierarchical Contrastive Alignment Loss (HCAL) to comprehensively align the video and text query. Extensive experiments on two benchmarks demonstrate that our method performs favorably against SoTA approaches. We will release all the source codes.

cs.CV↗

arXiv2025

HERO: Human Reaction Generation from Videos

Human reaction generation represents a significant research domain for interactive AI, as humans constantly interact with their surroundings. Previous works focus mainly on synthesizing the reactive motion given a human motion sequence. This paradigm limits interaction categories to human-human interactions and ignores emotions that may influence reaction generation. In this work, we propose to generate 3D human reactions from RGB videos, which involves a wider range of interaction categories and naturally provides information about expressions that may reflect the subject's emotions. To cope with this task, we present HERO, a simple yet powerful framework for Human rEaction geneRation from videOs. HERO considers both global and frame-level local representations of the video to extract the interaction intention, and then uses the extracted interaction intention to guide the synthesis of the reaction. Besides, local visual representations are continuously injected into the model to maximize the exploitation of the dynamic properties inherent in videos. Furthermore, the ViMo dataset containing paired Video-Motion data is collected to support the task. In addition to human-human interactions, these video-motion pairs also cover animal-human interactions and scene-human interactions. Extensive experiments demonstrate the superiority of our methodology. The code and dataset will be publicly available at https://jackyu6.github.io/HERO.

cs.CV↗

arXiv2025

QMambaBSR: Burst Image Super-Resolution with Query State Space Model

Burst super-resolution aims to reconstruct high-resolution images with higher quality and richer details by fusing the sub-pixel information from multiple burst low-resolution frames. In BusrtSR, the key challenge lies in extracting the base frame's content complementary sub-pixel details while simultaneously suppressing high-frequency noise disturbance. Existing methods attempt to extract sub-pixels by modeling inter-frame relationships frame by frame while overlooking the mutual correlations among multi-current frames and neglecting the intra-frame interactions, leading to inaccurate and noisy sub-pixels for base frame super-resolution. Further, existing methods mainly employ static upsampling with fixed parameters to improve spatial resolution for all scenes, failing to perceive the sub-pixel distribution difference across multiple frames and cannot balance the fusion weights of different frames, resulting in over-smoothed details and artifacts. To address these limitations, we introduce a novel Query Mamba Burst Super-Resolution (QMambaBSR) network, which incorporates a Query State Space Model (QSSM) and Adaptive Up-sampling module (AdaUp). Specifically, based on the observation that sub-pixels have consistent spatial distribution while random noise is inconsistently distributed, a novel QSSM is proposed to efficiently extract sub-pixels through inter-frame querying and intra-frame scanning while mitigating noise interference in a single step. Moreover, AdaUp is designed to dynamically adjust the upsampling kernel based on the spatial distribution of multi-frame sub-pixel information in the different burst scenes, thereby facilitating the reconstruction of the spatial arrangement of high-resolution details. Extensive experiments on four popular synthetic and real-world benchmarks demonstrate that our method achieves a new state-of-the-art performance.

cs.CV↗

arXiv2025

WeGen: A Unified Model for Interactive Multimodal Generation as We Chat

Existing multimodal generative models fall short as qualified design copilots, as they often struggle to generate imaginative outputs once instructions are less detailed or lack the ability to maintain consistency with the provided references. In this work, we introduce WeGen, a model that unifies multimodal generation and understanding, and promotes their interplay in iterative generation. It can generate diverse results with high creativity for less detailed instructions. And it can progressively refine prior generation results or integrating specific contents from references following the instructions in its chat with users. During this process, it is capable of preserving consistency in the parts that the user is already satisfied with. To this end, we curate a large-scale dataset, extracted from Internet videos, containing rich object dynamics and auto-labeled dynamics descriptions by advanced foundation models to date. These two information are interleaved into a single sequence to enable WeGen to learn consistency-aware generation where the specified dynamics are generated while the consistency of unspecified content is preserved aligned with instructions. Besides, we introduce a prompt self-rewriting mechanism to enhance generation diversity. Extensive experiments demonstrate the effectiveness of unifying multimodal understanding and generation in WeGen and show it achieves state-of-the-art performance across various visual generation benchmarks. These also demonstrate the potential of WeGen as a user-friendly design copilot as desired. The code and models will be available at https://github.com/hzphzp/WeGen.

cs.CV↗

arXiv2025

Pixel to Gaussian: Ultra-Fast Continuous Super-Resolution with 2D Gaussian Modeling

Arbitrary-scale super-resolution (ASSR) aims to reconstruct high-resolution (HR) images from low-resolution (LR) inputs with arbitrary upsampling factors using a single model, addressing the limitations of traditional SR methods constrained to fixed-scale factors (\textit{e.g.}, $\times$ 2). Recent advances leveraging implicit neural representation (INR) have achieved great progress by modeling coordinate-to-pixel mappings. However, the efficiency of these methods may suffer from repeated upsampling and decoding, while their reconstruction fidelity and quality are constrained by the intrinsic representational limitations of coordinate-based functions. To address these challenges, we propose a novel ContinuousSR framework with a Pixel-to-Gaussian paradigm, which explicitly reconstructs 2D continuous HR signals from LR images using Gaussian Splatting. This approach eliminates the need for time-consuming upsampling and decoding, enabling extremely fast arbitrary-scale super-resolution. Once the Gaussian field is built in a single pass, ContinuousSR can perform arbitrary-scale rendering in just 1ms per scale. Our method introduces several key innovations. Through statistical ana

cs.CV↗

arXiv2025

Get In Video: Add Anything You Want to the Video

Video editing increasingly demands the ability to incorporate specific real-world instances into existing footage, yet current approaches fundamentally fail to capture the unique visual characteristics of particular subjects and ensure natural instance/scene interactions. We formalize this overlooked yet critical editing paradigm as "Get-In-Video Editing", where users provide reference images to precisely specify visual elements they wish to incorporate into videos. Addressing this task's dual challenges, severe training data scarcity and technical challenges in maintaining spatiotemporal coherence, we introduce three key contributions. First, we develop GetIn-1M dataset created through our automated Recognize-Track-Erase pipeline, which sequentially performs video captioning, salient instance identification, object detection, temporal tracking, and instance removal to generate high-quality video editing pairs with comprehensive annotations (reference image, tracking mask, instance prompt). Second, we present GetInVideo, a novel end-to-end framework that leverages a diffusion transformer architecture with 3D full attention to process reference images, condition videos, and masks simultaneously, maintaining temporal coherence, preserving visual identity, and ensuring natural scene interactions when integrating reference objects into videos. Finally, we establish GetInBench, the first comprehensive benchmark for Get-In-Video Editing scenario, demonstrating our approach's superior performance through extensive evaluations. Our work enables accessible, high-quality incorporation of specific real-world subjects into videos, significantly advancing personalized video editing capabilities.

cs.CV↗

arXiv2025

Towards Realistic Data Generation for Real-World Super-Resolution

Existing image super-resolution (SR) techniques often fail to generalize effectively in complex real-world settings due to the significant divergence between training data and practical scenarios. To address this challenge, previous efforts have either manually simulated intricate physical-based degradations or utilized learning-based techniques, yet these approaches remain inadequate for producing large-scale, realistic, and diverse data simultaneously. In this paper, we introduce a novel Realistic Decoupled Data Generator (RealDGen), an unsupervised learning data generation framework designed for real-world super-resolution. We meticulously develop content and degradation extraction strategies, which are integrated into a novel content-degradation decoupled diffusion model to create realistic low-resolution images from unpaired real LR and HR images. Extensive experiments demonstrate that RealDGen excels in generating large-scale, high-quality paired data that mirrors real-world degradations, significantly advancing the performance of popular SR models on various real-world benchmarks.

cs.CV↗

arXiv2025

VanGogh: A Unified Multimodal Diffusion-based Framework for Video Colorization

Video colorization aims to transform grayscale videos into vivid color representations while maintaining temporal consistency and structural integrity. Existing video colorization methods often suffer from color bleeding and lack comprehensive control, particularly under complex motion or diverse semantic cues. To this end, we introduce VanGogh, a unified multimodal diffusion-based framework for video colorization. VanGogh tackles these challenges using a Dual Qformer to align and fuse features from multiple modalities, complemented by a depth-guided generation process and an optical flow loss, which help reduce color overflow. Additionally, a color injection strategy and luma channel replacement are implemented to improve generalization and mitigate flickering artifacts. Thanks to this design, users can exercise both global and local control over the generation process, resulting in higher-quality colorized videos. Extensive qualitative and quantitative evaluations, and user studies, demonstrate that VanGogh achieves superior temporal consistency and color fidelity.Project page: https://becauseimbatman0.github.io/VanGogh.

cs.CV↗

arXiv2024

RAIN: Real-time Animation of Infinite Video Stream

Live animation has gained immense popularity for enhancing online engagement, yet achieving high-quality, real-time, and stable animation with diffusion models remains challenging, especially on consumer-grade GPUs. Existing methods struggle with generating long, consistent video streams efficiently, often being limited by latency issues and degraded visual quality over extended periods. In this paper, we introduce RAIN, a pipeline solution capable of animating infinite video streams in real-time with low latency using a single RTX 4090 GPU. The core idea of RAIN is to efficiently compute frame-token attention across different noise levels and long time-intervals while simultaneously denoising a significantly larger number of frame-tokens than previous stream-based methods. This design allows RAIN to generate video frames with much shorter latency and faster speed, while maintaining long-range attention over extended video streams, resulting in enhanced continuity and consistency. Consequently, a Stable Diffusion model fine-tuned with RAIN in just a few epochs can produce video streams in real-time and low latency without much compromise in quality or consistency, up to infinite long. Despite its advanced capabilities, the RAIN only introduces a few additional 1D attention blocks, imposing minimal additional burden. Experiments in benchmark datasets and generating super-long videos demonstrating that RAIN can animate characters in real-time with much better quality, accuracy, and consistency than competitors while costing less latency. All code and models will be made publicly available.

cs.CV↗

arXiv2024

Leverage Task Context for Object Affordance Ranking

Intelligent agents accomplish different tasks by utilizing various objects based on their affordance, but how to select appropriate objects according to task context is not well-explored. Current studies treat objects within the affordance category as equivalent, ignoring that object affordances vary in priority with different task contexts, hindering accurate decision-making in complex environments. To enable agents to develop a deeper understanding of the objects required to perform tasks, we propose to leverage task context for object affordance ranking, i.e., given image of a complex scene and the textual description of the affordance and task context, revealing task-object relationships and clarifying the priority rank of detected objects. To this end, we propose a novel Context-embed Group Ranking Framework with task relation mining module and graph group update module to deeply integrate task context and perform global relative relationship transmission. Due to the lack of such data, we construct the first large-scale task-oriented affordance ranking dataset with 25 common tasks, over 50k images and more than 661k objects. Experimental results demonstrate the feasibility of the task context based affordance learning paradigm and the superiority of our model over state-of-the-art models in the fields of saliency ranking and multimodal object detection. The source code and dataset will be made available to the public.

cs.CV↗

arXiv2024

Unveiling Hidden Details: A RAW Data-Enhanced Paradigm for Real-World Super-Resolution

Real-world image super-resolution (Real SR) aims to generate high-fidelity, detail-rich high-resolution (HR) images from low-resolution (LR) counterparts. Existing Real SR methods primarily focus on generating details from the LR RGB domain, often leading to a lack of richness or fidelity in fine details. In this paper, we pioneer the use of details hidden in RAW data to complement existing RGB-only methods, yielding superior outputs. We argue that key image processing steps in Image Signal Processing, such as denoising and demosaicing, inherently result in the loss of fine details in LR images, making LR RAW a valuable information source. To validate this, we present RealSR-RAW, a comprehensive dataset comprising over 10,000 pairs with LR and HR RGB images, along with corresponding LR RAW, captured across multiple smartphones under varying focal lengths and diverse scenes. Additionally, we propose a novel, general RAW adapter to efficiently integrate LR RAW data into existing CNNs, Transformers, and Diffusion-based Real SR models by suppressing the noise contained in LR RAW and aligning its distribution. Extensive experiments demonstrate that incorporating RAW data significantly enhances detail recovery and improves Real SR performance across ten evaluation metrics, including both fidelity and perception-oriented metrics. Our findings open a new direction for the Real SR task, with the dataset and code will be made available to support future research.

eess.IV↗

arXiv2024

$\text{S}^{3}$Mamba: Arbitrary-Scale Super-Resolution via Scaleable State Space Model

Arbitrary scale super-resolution (ASSR) aims to super-resolve low-resolution images to high-resolution images at any scale using a single model, addressing the limitations of traditional super-resolution methods that are restricted to fixed-scale factors (e.g., $\times2$, $\times4$). The advent of Implicit Neural Representations (INR) has brought forth a plethora of novel methodologies for ASSR, which facilitate the reconstruction of original continuous signals by modeling a continuous representation space for coordinates and pixel values, thereby enabling arbitrary-scale super-resolution. Consequently, the primary objective of ASSR is to construct a continuous representation space derived from low-resolution inputs. However, existing methods, primarily based on CNNs and Transformers, face significant challenges such as high computational complexity and inadequate modeling of long-range dependencies, which hinder their effectiveness in real-world applications. To overcome these limitations, we propose a novel arbitrary-scale super-resolution method, called $\text{S}^{3}$Mamba, to construct a scalable continuous representation space. Specifically, we propose a Scalable State Space Model (SSSM) to modulate the state transition matrix and the sampling matrix of step size during the discretization process, achieving scalable and continuous representation modeling with linear computational complexity. Additionally, we propose a novel scale-aware self-attention mechanism to further enhance the network's ability to perceive global important features at different scales, thereby building the $\text{S}^{3}$Mamba to achieve superior arbitrary-scale super-resolution. Extensive experiments on both synthetic and real-world benchmarks demonstrate that our method achieves state-of-the-art performance and superior generalization capabilities at arbitrary super-resolution scales.

cs.CV↗