CVPR 2025 Papers

We introduce OminiFlow, a novel generative model designed for any-to-any generation tasks such as text-to-image, text-to-audio, and audio-to-image synthesis. OminiFlow advances the rectified flow (RF) framework used in text-to-image models to handle the joint distribution of multiple modalities. It outperforms previous any-to-any models on a wide range of tasks, such as text-to-image and text-to-audio synthesis. Our work offers three key contributions: First, we extend RF to a multi-modal setting and introduce a novel guidance mechanism, enabling users to flexibly control the alignment between different modalities in the generated outputs. Second, we propose a novel architecture that extends the text-to-image MMDiT architecture of Stable Diffusion 3 and enables audio and text generation. The extended modules can be efficiently pretrained individually and merged with the vanilla text-to-image MMDiT for fine-tuning. Lastly, we conduct a comprehensive study on the design choices of rectified flow transformers for large-scale audio and text generation, providing valuable insights into optimizing performance across diverse modalities.

The advances in vision-language models (VLMs) have led to a growing interest in autonomous driving to leverage their strong reasoning capabilities. However, extending these capabilities from 2D to full 3D understanding is crucial for real-world applications. To address this challenge, we propose OmniDrive, a holistic vision-language dataset that aligns agent models with 3D driving tasks through counterfactual reasoning. This approach enhances decision-making by evaluating potential scenarios and their outcomes, similar to human drivers considering alternative actions. Our counterfactual-based synthetic data annotation process generates large-scale, high-quality datasets, providing denser supervision signals that bridge planning trajectories and language-based reasoning. Futher, we explore two advanced OmniDrive-Agent frameworks, namely Omni-L and Omni-Q, to assess the importance of vision-language alignment versus 3D perception, revealing critical insights into designing effective LLM-agents. Significant improvements on the DriveLM Q&A benchmark and nuScenes open-loop planning demonstrate the effectiveness of our dataset and methods.

Document content extraction is a critical task in computer vision, underpinning the data needs of large language models (LLMs) and retrieval-augmented generation (RAG) systems. Despite recent progress, current document parsing methods have not been fairly and comprehensively evaluated due to the narrow coverage of document types and the simplified, unrealistic evaluation procedures in existing benchmarks. To address these gaps, we introduce OmniDocBench, a novel benchmark featuring high-quality annotations across nine document sources, including academic papers, textbooks, and more challenging cases such as handwritten notes and densely typeset newspapers. OmniDocBench supports flexible, multi-level evaluations--ranging from an end-to-end assessment to the task-specific and attribute-based analysis--using 19 layout categories and 15 attribute labels. We conduct a thorough evaluation of both pipeline-based methods and end-to-end vision-language models, revealing their strengths and weaknesses across different document types. OmniDocBench sets a new standard for the fair, diverse, and fine-grained evaluation in document parsing. Dataset and code are available at https://github.com/opendatalab/OmniDocBench.

Prior works employing pixel-based Gaussian representation have demonstrated efficacy in feed-forward sparse-view reconstruction. However, such representation necessitates cross-view overlap for accurate depth estimation, and is challenged by object occlusions and frustum truncations. As a result, these methods require scene-centric data acquisition to maintain cross-view overlap and complete scene visibility to circumvent occlusions and truncations, which limits their applicability to scene-centric reconstruction. In contrast, in autonomous driving scenarios, a more practical paradigm is ego-centric reconstruction, which is characterized by minimal cross-view overlap and frequent occlusions and truncations. The limitations of pixel-based representation thus hinder the utility of prior works in this task. In light of this, this paper conducts an in-depth analysis of different representations, and introduces Omni-Gaussian representation with tailored network design to complement their strengths and mitigate their drawbacks. Experiments show that our method significantly surpasses state-of-the-art methods, pixelSplat and MVSplat, in ego-centric reconstruction, and achieves comparable performance to prior works in scene-centric reconstruction. Our code is available at https://github.com/WU-CVGL/Omni-Scene.

We present Omni-RGPT, a multimodal large language model designed to facilitate region-level comprehension for both images and videos. To achieve consistent region representation across spatio-temporal dimensions, we introduce Token Mark, a set of tokens highlighting the target regions within the visual feature space. These tokens are directly embedded into spatial regions using region prompts (e.g., boxes or masks) and simultaneously incorporated into the text prompt to specify the target, establishing a direct connection between visual and text tokens. To further support robust video understanding without requiring tracklets, we introduce an auxiliary task that guides Token Mark by leveraging the consistency of the tokens, enabling stable region interpretation across the video. Additionally, we introduce a large-scale region-level video instruction dataset (RegVID-300k). Omni-RGPT achieves state-of-the-art results on image and video-based commonsense reasoning benchmarks while showing strong performance in captioning and referring expression comprehension tasks.

We introduce Omni-ID, a novel facial representation designed specifically for generative tasks. Omni-ID encodes holistic information about an individual's appearance across diverse expressions and poses within a fixed-size representation. It consolidates information from a varied number of unstructured input images into a structured representation, where each entry represents certain global or local identity features. Our approach uses a few-to-many identity reconstruction training paradigm, where a limited set of input images is used to reconstruct multiple target images of the same individual in various poses and expressions. A multi-decoder framework is introduced to leverage the complementary strengths of diverse decoders during training. Unlike conventional representations, such as ArcFace and CLIP, which are typically learned through discriminative or contrastive objectives, Omni-ID is optimized with a generative objective, resulting in a more comprehensive and nuanced identity capture for generative tasks. Trained on our MFHQ dataset -- a multi-view facial image collection, Omni-ID demonstrates substantial improvements over conventional representations across various generative tasks.

Neural surface reconstruction has been dominated by implicit representations with marching cubes for explicit surface extraction. However, those methods typically require high-quality normals for accurate reconstruction. We propose OffsetOPT, a method that reconstructs explicit surfaces directly from 3D point clouds and eliminates the need for point normals. The approach comprises two stages: first, we train a neural network to predict surface triangles based on local point geometry, given uniformly distributed training point clouds. Next, we apply the frozen network to reconstruct surfaces from unseen point clouds by optimizing a per-point offset to maximize the accuracy of triangle predictions. Compared to state-of-the-art methods, OffsetOPT not only excels at reconstructing overall surfaces but also significantly preserves sharp surface features. We demonstrate its accuracy on popular benchmarks, including small-scale shapes and large-scale open surfaces.

This paper introduces a novel anomaly detection (AD) problem aimed at identifying `odd-looking' objects within a scene by comparing them to other objects present. Unlike traditional AD benchmarks with fixed anomaly criteria, our task detects anomalies specific to each scene by inferring a reference group of regular objects. To address occlusions, we use multiple views of each scene as input, construct 3D object-centric models for each instance from 2D views, enhancing these models with geometrically consistent part-aware representations. Anomalous objects are then detected through cross-instance comparison. We also introduce two new benchmarks, ToysAD-8K and PartsAD-15K as testbeds for future research in this task. We provide a comprehensive analysis of our method quantitatively and qualitatively on these benchmarks.

Recent open-world representation learning approaches have leveraged CLIP to enable zero-shot 3D object recognition. However, performance on real point clouds with occlusions still falls short due to unrealistic pretraining settings. Additionally, these methods incur high inference costs because they rely on Transformer's attention modules. In this paper, we make two contributions to address these limitations. First, we propose occlusion-aware text-image-point cloud pretraining to reduce the training-testing domain gap. From 52K synthetic 3D objects, our framework generates nearly 630K partial point clouds for pretraining, consistently improving real-world recognition performances of existing popular 3D networks. Second, to reduce computational requirements, we introduce DuoMamba, a two-stream linear state space model tailored for point clouds. By integrating two space-filling curves with 1D convolutions, DuoMamba effectively models spatial dependencies between point tokens, offering a powerful alternative to Transformer. When pretrained with our framework, DuoMamba surpasses current state-of-the-art methods while reducing latency and FLOPs, highlighting the potential of our approach for real-world applications. Our code and data are available at ndkhanh360.github.io/project_occtip.

Training deep learning models for semantic occupancy prediction is challenging due to factors such as a large number of occupancy cells, severe occlusion, limited visual cues, complicated driving scenarios, etc. Recent methods often adopt transformer-based architectures given their strong capability in learning input-conditioned weights and long-range relationships. However, transformer-based networks are notorious for their quadratic computation complexity, seriously undermining their efficacy and deployment in semantic occupancy prediction. Inspired by the global modeling and linear computation complexity of the Mamba architecture, we present the first Mamba-based network for semantic occupancy prediction, termed OccMamba. Specifically, we first design the hierarchical Mamba module and local context processor to better aggregate global and local contextual information, respectively. Besides, to relieve the inherent domain gap between the linguistic and 3D domains, we present a simple yet effective 3D-to-1D reordering scheme, i.e., height-prioritized 2D Hilbert expansion. It can maximally retain the spatial structure of 3D voxels as well as facilitate the processing of Mamba blocks. Endowed with the aforementioned designs, our OccMamba is capable of directly and efficiently processing large volumes of dense scene grids, achieving state-of-the-art performance across three prevalent occupancy prediction benchmarks, including OpenOccupancy, SemanticKITTI, and SemanticPOSS. Notably, on OpenOccupancy, our OccMamba outperforms the previous state-of-the-art Co-Occ by 5.1% IoU and 4.3% mIoU, respectively. Our implementation is open-sourced and available at: https://github.com/USTCLH/OccMamba.

Simple as it seems, moving an object to another location within an image is, in fact, a challenging image-editing task that requires re-harmonizing the lighting, adjusting the pose based on perspective, accurately filling occluded regions, and ensuring coherent synchronization of shadows and reflections while maintaining the object identity. In this paper, we present ObjectMover, a generative model that can perform object movement in highly challenging scenes. Our key insight is that we model this task as a sequence-to-sequence problem and fine-tune a video generation model to leverage its knowledge of consistent object generation across video frames. We show that with this approach, our model is able to adjust to complex real-world scenarios, handling extreme lighting harmonization and object effect movement. As large-scale data for object movement are unavailable, we construct a data generation pipeline using a modern game engine to synthesize high-quality data pairs. We further propose a multi-task learning strategy that enables training on real-world video data to improve the model generalization. Through extensive experiments, we demonstrate that ObjectMover achieves outstanding results and adapts well to real-world scenarios.

Audio-visual sound source localization task aims to spatially localize sound-making objects within visual scenes by integrating visual and audio cues. However, existing methods struggle with accurately localizing sound-making objects in complex scenes, particularly when visually similar silent objects coexist. This limitation arises primarily from their reliance on simple audio-visual correspondence, which does not capture fine-grained semantic differences between sound-making and silent objects. To address these challenges, we propose a novel sound source localization framework leveraging Multimodal Large Language Models (MLLMs) to generate detailed contextual information that explicitly distinguishes between sound-making foreground objects and silent background objects. To effectively integrate this detailed information, we introduce two novel loss functions: Object-aware Contrastive Alignment (OCA) loss and Object Region Isolation (ORI) loss. Extensive experimental results on MUSIC and VGGSound datasets demonstrate the effectiveness of our approach, significantly outperforming existing methods in both single-source and multi-source localization scenarios. Code and generated detailed contextual information are available at: https://github.com/VisualAIKHU/OA-SSL.

Egocentric video grounding is a crucial task for embodied intelligence applications, distinct from exocentric video moment localization. Existing methods primarily focus on the distributional differences between egocentric and exocentric videos but often neglect key characteristics of egocentric videos and the fine-grained information emphasized by question-type queries. To address these limitations, we propose OSGNet, an Object-Shot enhanced Grounding Network for egocentric video. Specifically, we extract object information from videos to enrich video representation, particularly for objects highlighted in the textual query but not directly captured in the video features. Additionally, we analyze the frequent shot movements inherent to egocentric videos, leveraging these features to extract the wearer's attention information, which enhances the model's ability to perform modality alignment. Experiments conducted on three datasets demonstrate that OSGNet achieves state-of-the-art performance, validating the effectiveness of our approach. Our code can be found at https://github.com/Yisen-Feng/OSGNet.

In robotic manipulation, task goals can be conveyed through various modalities, such as language, goal images, and goal videos. However, natural language can be ambiguous, while images or videos may offer overly detailed specifications. To address these challenges, we propose a novel approach using comprehensive multi-modal prompts that explicitly convey both low-level actions and high-level planning in a simple manner. Specifically, for each key-frame in the task sequence, our method allows for manual or automatic generation of simple and expressive 2D visual prompts overlaid on RGB images. These prompts represent the required task goals, such as the end-effector pose and the desired movement direction after contact. We develop a training strategy that enables the model to interpret these visual-language prompts and predict the corresponding contact poses and movement directions in SE(3) space. Furthermore, by sequentially executing all key-frame steps, the model can complete long-horizon tasks. This approach not only helps the model explicitly understand the task objectives but also enhances its robustness on unseen tasks by providing easily interpretable prompts. We evaluate our method in both simulated and real-world environments, demonstrating its robust manipulation capabilities.

Object detection in event streams has emerged as a cutting-edge research area, demonstrating superior performance in low-light conditions, scenarios with motion blur, and rapid movements. Current detectors leverage spiking neural networks, Transformers, or convolutional neural networks as their core architectures, each with its own set of limitations including restricted performance, high computational overhead, or limited local receptive fields. This paper introduces a novel MoE (Mixture of Experts) heat conduction-based object detection algorithm that strikingly balances accuracy and computational efficiency. Initially, we employ a stem network for event data embedding, followed by processing through our innovative MoE-HCO blocks. Each block integrates various expert modules to mimic heat conduction within event streams. Subsequently, an IoU-based query selection module is utilized for efficient token extraction, which is then channeled into a detection head for the final object detection process. Furthermore, we are pleased to introduce EvDET200K, a novel benchmark dataset for event-based object detection. Captured with a high-definition Prophesee EVK4-HD event camera, this dataset encompasses 10 distinct categories, 200,000 bounding boxes, and 10,054 samples, each spanning 2 to 5 seconds. We also provide comprehensive results from over 15 state-of-the-art detectors, offering a solid foundation for future research and comparison.

Open world perception expands traditional closed-set frameworks, which assume a predefined set of known categories, to encompass dynamic real-world environments. Open World Object Detection (OWOD) and Open Vocabulary Object Detection (OVD) are two main research directions, each addressing unique challenges in dynamic environments. However, existing studies often focus on only one of these tasks, leaving the combined challenges of OWOD and OVD largely underexplored. In this paper, we propose a novel detector, OW-OVD, which inherits the zero-shot generalization capability of OVD detectors while incorporating the ability to actively detect unknown objects and progressively optimize performance through incremental learning, as seen in OWOD detectors. To achieve this, we start with a standard OVD detector and adapt it for OWOD tasks. For attribute selection, we propose the Visual Similarity Attribute Selection (VSAS) method, which identifies the most generalizable attributes by computing similarity distributions across annotated and unannotated regions. Additionally, to ensure the diversity of attributes, we incorporate a similarity constraint in the iterative process. Finally, to preserve the standard inference process of OVD, we propose the Hybrid Attribute-Uncertainty Fusion (HAUF) method. This method combines attribute similarity with known class uncertainty to infer the likelihood of an object belonging to an unknown class. We validated the effectiveness of OW-OVD through evaluations on two OWOD benchmarks, M-OWODB and S-OWODB. The results demonstrate that OW-OVD outperforms existing state-of-the-art models, achieving a +15.3 improvement in unknown object recall (U-Recall) and a +15.5 increase in unknown class average precision (U-mAP). Our code is available at: https://github.com/xxyzll/OW_OVD.

Temporal Awareness, the ability to reason dynamically based on the timestamp when a question is raised, is the key distinction between offline and online video LLMs. Unlike offline models, which rely on complete videos for static, post hoc analysis, online models process video streams incrementally and dynamically adapt their responses based on the timestamp at which the question is posed. Despite its significance, temporal awareness has not been adequately evaluated in existing benchmarks. To fill this gap, we present OVO-Bench (Online-VideO-Benchmark), a novel video benchmark that emphasizes the importance of timestamps for advanced online video understanding capability benchmarking. OVO-Bench evaluates the ability of video LLMs to reason and respond to events occurring at specific timestamps under three distinct scenarios: (1) Backward tracing: trace back to past events to answer the question. (2) Real-time understanding: understand and respond to events as they unfold at the current timestamp. (3) Forward active responding: delay the response until sufficient future information becomes available to answer the question accurately. OVO-Bench comprises 12 tasks, featuring 644 unique videos and approximately human-curated 2,800 fine-grained meta-annotations with precise timestamps. We combine automated generation pipelines with human curation. With these high-quality samples, we further developed an evaluation pipeline to systematically query video LLMs along the video timeline. Evaluations of nine Video-LLMs reveal that, despite advancements on traditional benchmarks, current models struggle with online video understanding, showing a significant gap compared to human agents. We hope OVO-Bench will drive progress in video LLMs and inspire future research in online video reasoning. Our benchmark and code can be accessed at https://joeleelyf.github.io/OVO-Bench.

Video diffusion models have shown great potential in generating high-quality videos, making them an increasingly popular focus. However, their inherent iterative nature leads to substantial computational and time costs. Although techniques such as consistency distillation and adversarial training have been employed to accelerate video diffusion by reducing inference steps, these methods often simply transfer the generation approaches from Image diffusion models to video diffusion models. As a result, these methods frequently fall short in terms of both performance and training stability. In this work, we introduce a two-stage training framework that effectively combines consistency distillation with adversarial training to address these challenges. Additionally, we propose a novel video discriminator design, which eliminates the need for decoding the video latents and improves the final performance. Our model is capable of producing high-quality videos in merely one-step, with the flexibility to perform multi-step refinement for further performance enhancement. Our quantitative evaluation on the OpenVid-1M benchmark shows that our model significantly outperforms existing methods. Notably, our 1-step performance (FVD 171.15) exceeds the 8-step performance of the consistency distillation based method, AnimateLCM (FVD 184.79), and approaches the 25-step performance of advanced Stable Video Diffusion (FVD 156.94).