Diffusion models have gradually gained prominence in the field of image synthesis, showcasing remarkable generative capabilities. Nevertheless, the slow inference and complex networks, resulting from redundancy at both temporal and structural levels, hinder their low-latency applications in real-world scenarios. Current acceleration methods for diffusion models focus separately on temporal and structural levels. However, independent optimization at each level to further push the acceleration limits results in significant performance degradation. On the other hand, integrating optimizations at both levels can compound the acceleration effects. Unfortunately, we find that the optimizations at these two levels are not entirely orthogonal. Performing separate optimizations and then simply integrating them results in unsatisfactory performance. To tackle this issue, we propose CacheQuant, a novel training-free paradigm that comprehensively accelerates diffusion models by jointly optimizing model caching and quantization techniques. Specifically, we employ a dynamic programming approach to determine the optimal cache schedule, in which the properties of caching and quantization are carefully considered to minimize errors. Additionally, we propose decoupled error correction to further mitigate the coupled and accumulated errors step by step. Experimental results show that CacheQuant achieves a 5.18xspeedup and 4xcompression for Stable Diffusion on MS-COCO, with only a 0.02 loss in CLIP score.
论文检索
输入标题、作者或关键词，从 9,654 篇学术成果中精准定位
Exploring real-world spaces using novel-view synthesis is fun, and reimagining those worlds in a different style adds another layer of excitement. Stylized worlds can also be used for downstream tasks where there is limited training data and a need to expand a model's training distribution. Most current novel-view synthesis stylization techniques lack the ability to convincingly change geometry. This is because any geometry change requires increased style strength which is often capped for stylization stability and consistency. In this work, we propose a new autoregressive 3D Gaussian Splatting stylization method. As part of this method, we contribute a new RGBD diffusion model that allows for strength control over appearance and shape stylization. To ensure consistency across stylized frames, we use a combination of novel depth-guided cross attention, feature injection, and a Warp ControlNet conditioned on composite frames for guiding the stylization of new frames. We validate our method via extensive qualitative results, quantitative experiments, and a user study.
Human motion synthesis aims to generate plausible human motion sequences, which has raised widespread attention in computer animation. Recent score-based generative models (SGMs) have demonstrated impressive results on this task. However, their training process involves complex curvature trajectories, leading to unstable training process.In this paper, we propose a Deterministic-to-Stochastic Diverse Latent Feature Mapping (DSDFM) method for human motion synthesis.DSDFM consists of two stages. The first human motion reconstruction stage aims to learn the latent space distribution of human motions. The second diverse motion generation stage aims to build connections between the Gaussian distribution and the latent space distribution of human motions, thereby enhancing the diversity and accuracy of the generated human motions. This stage is achieved by the designed deterministic feature mapping procedure with DerODE and stochastic diverse output generation procedure with DivSDE. DSDFM is easy to train compared to previous SGMs-based methods and can enhance diversity without introducing additional training parameters.Through qualitative and quantitative experiments, DSDFM achieves state-of-the-art results surpassing the latest methods, validating its superiority in human motion synthesis.
Recently, 3D Gaussian Splatting (3DGS) has enabled photorealistic view synthesis at high inference speeds. However, its splatting-based rendering model makes several approximations to the rendering equation, reducing physical accuracy. We show that splatting and its approximations are unnecessary, even within a rasterizer; we instead volumetrically integrate 3D Gaussians directly to compute the transmittance across them analytically. We use this analytic transmittance to derive more physically accurate alpha values than 3DGS, which can directly be used within their framework. The result is a method that more closely follows the volume rendering equation (similar to ray tracing) while enjoying the speed benefits of rasterization. Our method represents opaque surfaces with higher accuracy and fewer points than 3DGS. This enables it to outperform 3DGS for view synthesis (measured in SSIM and LPIPS). Being volumetrically consistent also enables our method to work out of the box for tomography. We match the state-of-the-art 3DGS-based tomography method with fewer points. Our code is publicly available at: https://github.com/chinmay0301ucsd/Vol3DGS
Recent advancements in Diffusion Transformer (DiT) have demonstrated remarkable proficiency in producing high-quality video content. Nonetheless, the potential of transformer-based diffusion models for effectively generating videos with controllable motion remains an area of limited exploration. This paper introduces Tora, the first trajectory-oriented DiT framework that concurrently integrates textual, visual, and trajectory conditions, thereby enabling scalable video generation with effective motion guidance. Specifically, Tora consists of a Trajectory Extractor (TE), a Spatial-Temporal DiT, and a Motion-guidance Fuser (MGF). The TE encodes arbitrary trajectories into hierarchical spacetime motion patches with a 3D motion compression network. The MGF integrates the motion patches into the DiT blocks to generate consistent videos that accurately follow designated trajectories. Our design aligns seamlessly with DiT's scalability, allowing precise control of video content's dynamics with diverse durations, aspect ratios, and resolutions. Extensive experiments demonstrate that Tora excels in achieving high motion fidelity compared to the foundational DiT model, while also accurately simulating the complex movements of the physical world. Code is made available at https://github.com/alibaba/Tora .
OnlineAnySeg: Online Zero-Shot 3D Segmentation by Visual Foundation Model Guided 2D Mask Merging
PDF ↗Online 3D open-vocabulary segmentation of a progressively reconstructed scene is both a critical and challenging task for embodied applications. With the success of visual foundation models (VFMs) in the image domain, leveraging 2D priors to address 3D online segmentation has become a prominent research focus. Since segmentation results provided by 2D priors often require spatial consistency to be lifted into final 3D segmentation, an efficient method for identifying spatial overlap among 2D masks is essential--yet existing methods rarely achieve this in real time, mainly limiting its use to offline approaches. To address this, we propose an efficient method that lifts 2D masks generated by VFMs into a unified 3D instance using a hashing technique. By employing voxel hashing for efficient 3D scene querying, our approach reduces the time complexity of costly spatial overlap queries from O(n^2) to O(n). Accurate spatial associations further enable 3D merging of 2D masks through simple similarity-based filtering in a zero-shot manner, making our approach more robust to incomplete and noisy data. Evaluated on the ScanNet and SceneNN benchmarks, our approach achieves state-of-the-art performance in online, open-vocabulary 3D instance segmentation with leading efficiency.
While Structure-from-Motion (SfM) has seen much progress over the years, state-of-the-art systems are prone to failure when facing extreme viewpoint changes in low-overlap, low-parallax or high-symmetry scenarios. Because capturing images that avoid these pitfalls is challenging, this severely limits the wider use of SfM, especially by non-expert users. We overcome these limitations by augmenting the classical SfM paradigm with monocular depth and normal priors inferred by deep neural networks. Thanks to a tight integration of monocular and multi-view constraints, our approach significantly outperforms existing ones under extreme viewpoint changes, while maintaining strong performance in standard conditions. We also show that monocular priors can help reject faulty associations due to symmetries, which is a long-standing problem for SfM. This makes our approach the first capable of reliably reconstructing challenging indoor environments from few images. Through principled uncertainty propagation, it is robust to errors in the priors, can handle priors inferred by different models with little tuning, and will thus easily benefit from future progress in monocular depth and normal estimation. Our code is publicly available at https://github.com/cvg/mpsfm.
Feedback is essential for learning a new skill or improving one's current skill-level. However, current methods for skill-assessment from video only provide scores or compare demonstrations, leaving the burden of knowing what to do differently on the user. We introduce a novel method to generate actionable feedback (AF) from video of a person doing a physical activity, such as basketball or soccer. Our method takes a video demonstration and its accompanying 3D body pose and generates (1) free-form expert commentary describing what the person is doing well and what they could improve, and (2) a visual expert demonstration that incorporates the required corrections. We show how to leverage Ego-Exo4D's videos of skilled activity and expert commentary together with a strong language model to create a weakly-supervised training dataset for this task, and we devise a multimodal video-language model to infer coaching feedback. Our method is able to reason across multi-modal input combinations to output full-spectrum, actionable coaching--expert commentary, expert video retrieval, and expert pose generation--outperforming strong vision-language models on both established metrics and human preference studies.
We present Articulated Kinematics Distillation (AKD), a framework for generating high-fidelity character animations by merging the strengths of skeleton-based animation and modern generative models. AKD uses a skeleton-based representation for rigged 3D assets, drastically reducing the Degrees of Freedom (DoFs) by focusing on joint-level control, which allows for efficient, consistent motion synthesis. Through Score Distillation Sampling (SDS) with pre-trained video diffusion models, AKD distills complex, articulated motions while maintaining structural integrity, overcoming challenges faced by 4D neural deformation fields in preserving shape consistency. This approach is naturally compatible with physics-based simulation, ensuring physically plausible interactions. Experiments show that AKD achieves superior 3D consistency and motion quality compared with existing works on text-to-4D generation.
Detection-Friendly Nonuniformity Correction: A Union Framework for Infrared UAV Target Detection
PDF ↗Infrared unmanned aerial vehicle (UAV) images captured using thermal detectors are often affected by temperature-dependent low-frequency nonuniformity, which significantly reduces the contrast of the images. Detecting UAV targets under nonuniform conditions is crucial in UAV surveillance applications. Existing methods typically treat infrared nonuniformity correction (NUC) as a preprocessing step for detection, which leads to suboptimal performance. Balancing the two tasks while enhancing detection-beneficial information remains challenging. In this paper, we present a detection-friendly union framework, termed UniCD, that simultaneously addresses both infrared NUC and UAV target detection tasks in an end-to-end manner. We first model NUC as a small number of parameter estimation problem jointly driven by priors and data to generate detection-conducive images. Then, we incorporate a new auxiliary loss with target mask supervision into the backbone of the infrared UAV target detection network to strengthen target features while suppressing the background. To better balance correction and detection, we introduce a detection-guided self-supervised loss to reduce feature discrepancies between the two tasks, thereby enhancing detection robustness to varying nonuniformity levels. Additionally, we construct a new benchmark composed of 50,000 infrared images in various nonuniformity types, multi-scale UAV targets and rich backgrounds with target annotations, called IRBFD. Extensive experiments on IRBFD demonstrate that our UniCD is a robust union framework for NUC and UAV target detection while achieving real-time processing capabilities. Dataset can be available at https://github.com/IVPLaboratory/UniCD.
While multimodal large language models demonstrate strong performance in complex reasoning tasks, they pose significant challenges related to model complexity during deployment, especially for resource-limited devices. In this paper, we propose an automatic pruning method for large vision-language models to enhance the efficiency of multimodal reasoning. Conventional methods rely on the training data of the original model to select the proper pruning ratio for different network components. However, these methods are impractical for large vision-language models due to the unaffordable search costs caused by web-scale training corpus. In contrast, our approach only leverages a small number of samples to search for the desired pruning policy by maximizing its generalization ability on unknown training data while maintaining the model accuracy, which enables the achievement of an optimal trade-off between accuracy and efficiency for large visual language models. Specifically, we formulate the generalization gap of the pruning strategy using the structural risk minimization principle. Based on both task performance and generalization capability, we iteratively search for the optimal pruning policy within a given search space and optimize the vision projector to evolve the search space with higher upper bound of performance. We conduct extensive experiments on the ScienceQA, Vizwiz, MM-vet, and LLaVA-Bench datasets for the task of visual question answering. Using only 64 samples for pruning policy search, EfficientLLaVA achieves an accuracy of 83.05% on ScienceQA, along with a x1.8 speedup compared to the dense LLaVA-v1.5-7B model.
Precipitation nowcasting involves using current radar observation sequences to predict future radar sequences and determine future precipitation distribution, which is crucial for disaster warning, traffic planning, and agricultural production. Despite numerous advancements, challenges persist in accurately predicting both the location and intensity of precipitation, as these factors are often interdependent, with complex atmospheric dynamics and moisture distribution causing position and intensity changes to be intricately coupled. Inspired by the fact that in the frequency domain, phase variations are shown to correspond to changes in the position of precipitation, while amplitude variations are linked to intensity changes, we propose an amplitude-phase disentanglement model called AlphaPre, which separately learn the position and intensity changes of precipitation. AlphaPre comprises three key components: a phase network, an amplitude network, and an AlphaMixer. The phase network captures positional changes by learning phase variations, while the amplitude network models intensity changes by alternating between the frequency and spatial domains. The AlphaMixer then integrates these components to produce a refined precipitation forecast. Extensive experiments on four datasets demonstrate the effectiveness and superiority of our method over state-of-the-art approaches. Our code is publicly available at https://github.com/linkenghong/AlphaPre.
Leveraging Global Stereo Consistency for Category-Level Shape and 6D Pose Estimation from Stereo Images
PDF ↗Stereo-based category-level shape and 6D pose estimation methods have the potential to generalize to a wider range of materials than RGBD methods, which often suffer from depth measurement errors. However, without explicit depth from two views, parameters to be estimated can become inherently entangled, negatively impacting performance. To address this, we propose a method that leverages global stereo consistency to constrain optimization directions and mitigate parameter entanglement. We first estimate an intra-category occupancy field to represent a unified shape across views, ensuring consistency and preventing shape ambiguity. Through a divide-and-conquer approach within global shape fitting, we fit this shape to stereo images to obtain the pose, iteratively rendering normalized depth maps and exchanging information across views. This approach improves convergence toward the correct pose and scale. We validated our method on both depth-friendly and depth-challenging materials using our S-RGBD dataset and the TOD benchmark. Our method surpasses RGBD methods on challenging objects and performs comparably on depth-friendly ones. Ablation studies confirm the effectiveness of each component.
Unlike modern native digital videos, the restoration of old films requires addressing specific degradations inherent to analog sources. However, existing specialized methods still fall short compared to general video restoration techniques. In this work, we propose a new baseline to re-examine the challenges in old film restoration. First, we develop an improved Mamba-based framework, dubbed MambaOFR, which can dynamically adjust the degradation removal patterns by generating degradation-aware prompts to tackle the complex and composite degradations present in old films. Second, we introduce a flow-guided mask deformable alignment module to mitigate the propagation of structured defect features in the temporal domain. Third, we introduce the first benchmark dataset that includes both synthetic and real-world old film clips. Extensive experiments show that the proposed method achieves state-of-the-art performance, outperforming existing advanced approaches in old film restoration. The implementation and model is available at https://github.com/MaoAYD/MambaOFR.
The Illusion of Unlearning: The Unstable Nature of Machine Unlearning in Text-to-Image Diffusion Models
PDF ↗Text-to-image models such as Stable Diffusion, DALL*E, and Midjourney have gained immense popularity lately. However, they are trained on vast amounts of data that may include private, explicit, or copyrighted material used without permission, raising serious legal and ethical concerns. In light of the recent regulations aimed at protecting individual data privacy, there has been a surge in Machine Unlearning methods designed to remove specific concepts from these models. However, we identify a critical flaw in these unlearning techniques: unlearned concepts will revive when the models are fine-tuned, even with general or unrelated prompts. In this paper, for the first time, through an extensive study, we demonstrate the unstable nature of existing unlearning methods in text-to-image diffusion models. We introduce a framework that includes a couple of measures for analyzing the stability of existing unlearning methods. Further, the paper offers preliminary insights into the plausible explanation for the instability of the mapping-based unlearning methods that can guide future research toward more robust unlearning techniques. Codes for implementing the proposed framework are provided.
MFogHub: Bridging Multi-Regional and Multi-Satellite Data for Global Marine Fog Detection and Forecasting
PDF ↗Deep learning approaches for marine fog detection and forecasting have outperformed traditional methods, demonstrating significant scientific and practical importance. However, the limited availability of open-source datasets remains a major challenge. Existing datasets, often focused on a single region or satellite, restrict the ability to evaluate model performance across diverse conditions and hinder the exploration of intrinsic marine fog characteristics. To address these limitations, we introduce MFogHub, the first multi-regional and multi-satellite dataset to integrate annotated marine fog observations from 15 coastal fog-prone regions and six geostationary satellites, comprising over 68,000 high-resolution samples. By encompassing diverse regions and satellite perspectives, MFogHub facilitates rigorous evaluation of both detection and forecasting methods under varying conditions. Extensive experiments with 16 baseline models demonstrate that MFogHub can reveal generalization fluctuations due to regional and satellite discrepancy, while also serving as a valuable resource for the development of targeted and scalable fog prediction techniques. Through MFogHub, we aim to advance both the practical monitoring and scientific understanding of marine fog dynamics on a global scale. The dataset and code are available in the supplementary materials.
Global effective receptive field plays a crucial role for image style transfer (ST) to obtain high-quality stylized results. However, existing ST backbones (e.g., CNNs and Transformers) suffer huge computational complexity to achieve global receptive fields. Recently, the State Space Model (SSM), especially the improved variant Mamba, has shown great potential for long-range dependency modeling with linear complexity, which offers a approach to resolve the above dilemma. In this paper, we develop a Mamba-based style transfer framework, termed SaMam. Specifically, a mamba encoder is designed to efficiently extract content and style information. In addition, a style-aware mamba decoder is developed to flexibly adapt to various styles. Moreover, to address the problems of local pixel forgetting, channel redundancy and spatial discontinuity of existing SSMs, we introduce both local enhancement and zigzag scan. Qualitative and quantitative results demonstrate that our SaMam outperforms state-of-the-art methods in terms of both accuracy and efficiency.
Environment understanding in egocentric videos is an important step for applications like robotics, augmented reality and assistive technologies. These videos are characterized by dynamic interactions and a strong dependence on the wearer's engagement with the environment. Traditional approaches often focus on isolated clips or fail to integrate rich semantic and geometric information, limiting scene comprehension. We introduce Dynamic Image-Video Feature Fields (DIV-FF), a framework that decomposes the egocentric scene into persistent, dynamic, and actor-based components while integrating both image and video-language features. Our model enables detailed segmentation, captures affordances, understands the surroundings and maintains consistent understanding over time. DIV-FF outperforms state-of-the-art methods, particularly in dynamically evolving scenarios, demonstrating its potential to advance long-term, spatio-temporal scene understanding.
Eyeglass reflection removal can restore the texture information in the reflection destructed eye area, which is meaningful for various tasks on the facial images. It is still challenging to correctly eliminate reflections, reasonably restore the lost contents, and guarantee that the final result has a consistent color and illumination with the input image. In this paper, we introduce a Degradation-guided Local-to-Global (DL2G) restoration framework to address this problem. We first propose a multiplicative reflection degradation model, which is used to alleviate reflection degradation to obtain a preliminary result. Then, in the local details restoration stage, we propose a local structure-aware diffusion model to learn the true distribution of texture details in the eye area. This helps in recovering lost contents in the regions of heavy degradation where the background is invisible. Finally, in the global consistency refinement stage, we utilize the input image as a reference image to generate the final result that is consistent with the input image in color and illumination. Extensive experiments demonstrate that our method can improve the effect of reflection removal and generate results with more reasonable semantics, exquisite details, and harmonious illumination.
Capturing the temporal evolution of Gaussian properties such as position, rotation, and scale is a challenging task due to the vast number of time-varying parameters and the limited photometric data available, which generally results in convergence issues, making it difficult to find an optimal solution. While feeding all inputs into an end-to-end neural network can effectively model complex temporal dynamics, this approach lacks explicit supervision and struggles to generate high-quality transformation fields. On the other hand, using time-conditioned polynomial functions to model Gaussian trajectories and orientations provides a more explicit and interpretable solution, but requires significant handcrafted effort and lacks generalizability across diverse scenes. To overcome these limitations, this paper introduces a novel approach based on a learnable infinite Taylor Formula to model the temporal evolution of Gaussians. This method offers both the flexibility of an implicit network-based approach and the interpretability of explicit polynomial functions, allowing for more robust and generalizable modeling of Gaussian dynamics across various dynamic scenes.Extensive experiments on dynamic novel view rendering tasks are conducted on public datasets, demonstrating that the proposed method achieves state-of-the-art performance in this domain.