Video restoration poses non-trivial challenges in maintaining fidelity while recovering temporally consistent details from unknown degradations in the wild. Despite recent advances in diffusion-based restoration, these methods often face limitations in generation capability and sampling efficiency. In this work, we present \text SeedVR , a diffusion transformer designed to handle real-world video restoration with arbitrary length and resolution. The core design of SeedVR lies in the shifted window attention that facilitates effective restoration on long video sequences. SeedVR further supports variable-sized windows near the boundary of both spatial and temporal dimensions, overcoming the resolution constraints of traditional window attention. Equipped with contemporary practices, including causal video autoencoder, mixed image and video training, and progressive training, SeedVR achieves highly-competitive performance on both synthetic and real-world benchmarks, as well as AI-generated videos. Extensive experiments demonstrate SeedVR's superiority over existing methods for generic video restoration.
论文检索
输入标题、作者或关键词，从 3,199 篇学术成果中精准定位
Recent works have shown that visual pretraining on egocentric datasets using masked autoencoders (MAE) can improve generalization for downstream robotics tasks. However, these approaches pretrain only on 2D images, while many robotics applications require 3D scene understanding. In this work, we propose 3D-MVP, a novel approach for 3D Multi-View Pretraining using masked autoencoders. We leverage Robotic View Transformer (RVT), which uses a multi-view transformer to understand the 3D scene and predict gripper pose actions. We split RVT's multi-view transformer into visual encoder and action decoder, and pretrain its visual encoder using masked autoencoding on large-scale 3D datasets such as Objaverse. We evaluate 3D-MVP on a suite of virtual robot manipulation tasks and demonstrate improved performance over baselines. Our results suggest that 3D-aware pretraining is a promising approach to improve generalization of vision-based robotic manipulation policies.
Multi-Modal Contrastive Masked Autoencoders: A Two-Stage Progressive Pre-training Approach for RGBD Datasets
PDF ↗In this paper, we propose a new progressive pre-training method for image understanding tasks which leverages RGB-D datasets. The method utilizes Multi-Modal Contrastive Masked Autoencoder and Denoising techniques. Our proposed approach consists of two stages. In the first stage, we pre-train the model using contrastive learning to learn cross-modal representations. In the second stage, we further pre-train the model using masked autoencoding and denoising/noise prediction used in diffusion models. Masked autoencoding focuses on reconstructing the missing patches in the input modality using local spatial correlations, while denoising learns high frequency components of the input data. Moreover, it incorporates global distillation in the second stage by leveraging the knowledge acquired in stage one. Our approach is scalable, robust and suitable for pre-training RGB-D datasets. Extensive experiments on multiple datasets such as ScanNet, NYUv2 and SUN RGB-D show the efficacy and superior performance of our approach. Specifically, we show an improvement of +1.3% mIoU against Mask3D on ScanNet semantic segmentation. We further demonstrate the effectiveness of our approach in low-data regime by evaluating it for semantic segmentation task against the state-of-the-art methods.
We present HuMoCon, a novel motion-video understanding framework designed for advanced human behavior analysis. The core of our method is a human motion concept discovery framework that efficiently trains multi-modal encoders to extract semantically meaningful and generalizable features. HuMoCon addresses key challenges in motion concept discovery for understanding and reasoning, including the lack of explicit multi-modality feature alignment and the loss of high-frequency information in masked autoencoding frameworks. Our approach integrates a feature alignment strategy that leverages video for contextual understanding and motion for fine-grained interaction modeling, further with a velocity reconstruction mechanism to enhance high-frequency feature expression and mitigate temporal over-smoothing. Comprehensive experiments on standard benchmarks demonstrate that HuMoCon enables effective motion concept discovery and significantly outperforms state-of-the-art methods in training large models for human motion understanding.
UHD-processer: Unified UHD Image Restoration with Progressive Frequency Learning and Degradation-aware Prompts
PDF ↗We introduce UHD-Processor, a unified and robust framework for all-in-one image restoration, which is particularly resource-efficient for Ultra-High-Definition (UHD) images. To address the limitations of traditional all-in-one methods that rely on complex restoration backbones, our strategy employs a frequency domain decoupling progressive learning technique, motivated by curriculum learning, to incrementally learn restoration mappings from low to high frequencies. This approach incorporates specialized sub-network modules to effectively tackle different frequency bands in a divide-and-conquer manner, significantly enhancing the learning capability of simpler networks. Moreover, to accommodate the high-resolution characteristics of UHD images, we developed a variational autoencoder (VAE)-based framework that reduces computational complexity by modeling a concise latent space. It integrates task-specific degradation awareness in the encoder and frequency selection in the decoder, enhancing task comprehension and generalization. Our unified model is able to handle various degradations such as denoising, deblurring, dehazing, low-lighting, etc. Experimental evaluations extensively showcase the effectiveness of our dual-strategy approach, significantly improving UHD image restoration and achieving cutting-edge performance across diverse conditions. The code will be available at https://github.com/lyd-2022/UHD-processer.
Concept bottleneck models (CBM) aim to produce inherently interpretable models that rely on human-understandable concepts for their predictions. However, existing approaches to design interpretable generative models based on CBMs are not yet efficient and scalable, as they require expensive generative model training from scratch as well as real images with labor-intensive concept supervision. To address these challenges, we present two novel and low-cost methods to build interpretable generative models through post-hoc techniques and we name our approaches: concept-bottleneck autoencoder (CB-AE) and concept controller (CC). Our proposed approaches enable efficient and scalable training without the need of real data and require only minimal to no concept supervision. Additionally, our methods generalize across modern generative model families including generative adversarial networks and diffusion models. We demonstrate the superior interpretability and steerability of our methods on numerous standard datasets like CelebA, CelebA-HQ, and CUB with large improvements (average 25%) over the prior work, while being 4-15x faster to train. Finally, a large-scale user study is performed to validate the interpretability and steerability of our methods.
Domain shift (the difference between source and target domains) poses a significant challenge in clinical applications, e.g., Diabetic Retinopathy (DR) grading. Despite considering certain clinical requirements, like source data privacy, conventional transfer methods are predominantly model-centered and often struggle to prevent model-targeted attacks. In this paper, we address a challenging Online Model-aGnostic Domain Adaptation (OMG-DA) setting, driven by the demands of clinical environments. This setting is characterized by the absence of the model and the flow of target data. To tackle the new challenge, we propose a novel approach, Generative Unadversarial ExampleS (GUES), which enables adaptation from a data-centric perspective. Specifically, we first theoretically reformulate conventional perturbation optimization in a generative way-- learning a perturbation generation function with a latent input variable. During model instantiation, we leverage a Variational AutoEncoder to express this function. The encoder with the reparameterization trick predicts the latent input, whilst the decoder is responsible for the generation. Furthermore, the saliency map is selected as pseudo perturbation labels. Because it not only captures potential lesions but also theoretically provides an upper bound on the function input, enabling the identification of the latent variable. Extensive experiments on DR benchmarks with both frozen pre-trained models and trainable models demonstrate the superiority of GUES, showing robustness even with small batch size. The source code and data are available at https://github.com/tntek/GUES.
Multi-layer image generation is a fundamental task that enables users to isolate, select, and edit specific image layers, thereby revolutionizing interactions with generative models. In this paper, we introduce the Anonymous Region Transformer (ART), which facilitates the direct generation of variable multi-layer transparent images based on a global text prompt and an anonymous region layout. Inspired by Schema theory, this anonymous region layout allows the generative model to autonomously determine which set of visual tokens should align with which text tokens, which is in contrast to the previously dominant semantic layout for the image generation task. In addition, the layer-wise region crop mechanism, which only selects the visual tokens belonging to each anonymous region, significantly reduces attention computation costs and enables the efficient generation of images with numerous distinct layers (e.g., 50+). When compared to the full attention approach, our method is over 12 times faster and exhibits fewer layer conflicts. Furthermore, we propose a high-quality multi-layer transparent image autoencoder that supports the direct encoding and decoding of the transparency of variable multi-layer images in a joint manner. By enabling precise control and scalable layer generation, ART establishes a new paradigm for interactive content creation.
Robust Multimodal Survival Prediction with Conditional Latent Differentiation Variational AutoEncoder
PDF ↗The integrative analysis of histopathological images and genomic data has received increasing attention for survival prediction of human cancers. However, the existing studies always hold the assumption that full modalities are available. As a matter of fact, the cost for collecting genomic data is high, which sometimes makes genomic data unavailable in testing samples. A common way of tackling such incompleteness is to generate the genomic representations from the pathology images. Nevertheless, such strategy still faces the following two challenges: (1) The gigapixel whole slide images (WSIs) are huge and thus hard for representation. (2) It is difficult to generate the genomic embeddings with diverse function categories in a unified generative framework. To address the above challenges, we propose a Conditional Latent Differentiation Variational AutoEncoder (LD-CVAE) for robust multimodal survival prediction, even with missing genomic data. Specifically, a Variational Information Bottleneck Transformer (VIB-Trans) module is proposed to learn compressed pathological representations from the gigapixel WSIs. To generate different functional genomic features, we develop a novel Latent Differentiation Variational AutoEncoder (LD-VAE) to learn the genomic and function-specific posteriors for the genomic embeddings with diverse functions. Finally, we use the product-of-experts technique to integrate the genomic posterior and image posterior for the joint latent distribution estimation in LD-CVAE. We test the effectiveness of our method on five different cancer datasets, and the experimental results demonstrate its superiority in both complete and missing modality scenarios.
While flourishing developments have been witnessed in text-to-motion generation, synthesizing physically realistic, controllable, language-conditioned Human Scene Interactions (HSI) remains a relatively underexplored landscape. Current HSI methods naively rely on conditional Variational AutoEncoder (cVAE) and diffusion models. They are typically associated with limited modalities of control signals and task-specific frameworks design, leading to inflexible adaptation across various interaction scenarios and descriptive-unfaithful motions in diverse 3D physical environments. In this paper, we propose HSI-GPT, a General-Purpose Large Scene-Motion-Language Model that applies "next-token prediction" paradigm of Large Language Models to the HSI domain. HSI-GPT not only exhibits remarkable flexibility to accommodate diverse control signals (3D scenes, textual commands, key-frame poses, as well as scene affordances), but it seamlessly supports various HSI-related tasks (e.g., multi-modal controlled HSI generation, HSI understanding, and general motion completion in 3D scenes). First, HSI-GPT quantizes textual descriptions and human motions into discrete, LLM-interpretable tokens with multi-modal tokenizers. Inspired by multi-modal learning, we develop a recipe for aligning mixed-modality tokens into the shared embedding space of LLMs. These interaction tokens are then organized into unified instruction following prompts, allowing HSI-GPT to fine-tune on question-and-answer tasks. Extensive experiments and visualizations validate that our general-purpose HSI-GPT model delivers exceptional performance across multiple HSI-related tasks.
Video Variational Autoencoder (VAE) encodes videos into a low-dimensional latent space, becoming a key component of most Latent Video Diffusion Models (LVDMs) to reduce model training costs. However, as the resolution and duration of generated videos increase, the encoding cost of Video VAEs becomes a limiting bottleneck in training LVDMs. Moreover, the block-wise inference method adopted by most LVDMs can lead to discontinuities of latent space when processing long-duration videos. The key to addressing the computational bottleneck lies in decomposing videos into distinct components and efficiently encoding the critical information. Wavelet transform can decompose videos into multiple frequency-domain components and improve the efficiency significantly, we thus propose Wavelet Flow VAE (WF-VAE), an autoencoder that leverages multi-level wavelet transform to facilitate low-frequency energy flow into latent representation. Furthermore, we introduce a method called Causal Cache, which maintains the integrity of latent space during block-wise inference. Compared to state-of-the-art video VAEs, WF-VAE demonstrates superior performance in both PSNR and LPIPS metrics, achieving 2x higher throughput and 4x lower memory consumption while maintaining competitive reconstruction quality. Our code and models are available at https://github.com/PKU-YuanGroup/WF-VAE.
Current research on generating 3D hand-object interaction motion primarily focuses on in-domain objects. Generalization to unseen objects is essential for practical applications, yet it remains both challenging and largely unexplored.In this paper, we propose LatentHOI, a novel approach designed to tackle the challenges of generalizing hand-object interaction to unseen objects.Our main insight lies in decoupling high-level temporal motion from fine-grained spatial hand-object interactions with a latent diffusion model coupled with a Grasping Variational Autoencoder (GraspVAE). This configuration not only enhances the conditional dependency between spatial grasp and temporal motion but also improves data utilization and reduces overfitting through regularization in the latent space. We conducted extensive experiments in an unseen-object setting on both single-hand grasping and bi-manual motion datasets, including GRAB, DexYCB, and OakInk.Quantitative and qualitative evaluations demonstrate that our method significantly enhances the realism and physical plausibility of generated motions for unseen objects, both in single and bimanual manipulations, compared to the state-of-the-art.
Recent advances in audio-visual learning have shown promising results in learning representations across modalities. However, most approaches rely on global audio representations that fail to capture fine-grained temporal correspondences with visual frames.Additionally, existing methods often struggle with conflicting optimization objectives when trying to jointly learn reconstruction and cross-modal alignment. In this work, we propose CAV-MAE Sync as a simple yet effective extension of the original CAV-MAE framework for self-supervised audio-visual learning. We address three key challenges: First, we tackle the granularity mismatch between modalities by treating audio as a temporal sequence aligned with video frames, rather than using global representations. Second, we resolve conflicting optimization goals by separating contrastive and reconstruction objectives through dedicated global tokens. Third, we improve spatial localization by introducing learnable register tokens that reduce semantic load on patch tokens. We evaluate the proposed approach on AudioSet, VGG Sound, and the ADE20K Sound dataset on zero-shot retrieval, classification and localization tasks demonstrating state-of-the-art performance and outperforming more complex architectures. Code available at https://github.com/edsonroteia/cav-mae-sync.
UniRestore: Unified Perceptual and Task-Oriented Image Restoration Model Using Diffusion Prior
PDF ↗Image restoration aims to recover content from inputs degraded by various factors, such as adverse weather, blur, and noise. Perceptual Image Restoration (PIR) methods improve visual quality but often do not support downstream tasks effectively. On the other hand, Task-oriented Image Restoration (TIR) methods focus on enhancing image utility for high-level vision tasks, sometimes compromising visual quality. This paper introduces UniRestore, a unified image restoration model that bridges the gap between PIR and TIR by using a diffusion prior. The diffusion prior is designed to generate images that align with human visual quality preferences, but these images are often unsuitable for TIR scenarios. To solve this limitation, UniRestore utilizes encoder features from an autoencoder to adapt the diffusion prior to specific tasks. We propose a Complementary Feature Restoration Module (CFRM) to reconstruct degraded encoder features and a Task Feature Adapter (TFA) module to facilitate adaptive feature fusion in the decoder. This design allows UniRestore to optimize images for both human perception and downstream task requirements, addressing discrepancies between visual quality and functional needs. Integrating these modules also enhances UniRestore's adapability and efficiency across diverse tasks. Extensive expertments demonstrate the superior performance of UniRestore in both PIR and TIR scenarios.
Autoregressive models have demonstrated remarkable success across various fields, from large language models (LLMs) to large multimodal models (LMMs) and 2D content generation, moving closer to artificial general intelligence (AGI). Despite these advances, applying autoregressive approaches to 3D object generation and understanding remains largely unexplored. This paper introduces Scale AutoRegressive 3D (SAR3D), a novel framework that leverages a multi-scale 3D vector-quantized variational autoencoder (VQVAE) to tokenize 3D objects for efficient autoregressive generation and detailed understanding. By predicting the next scale in a multi-scale latent representation instead of the next single token, SAR3D reduces generation time significantly, achieving fast 3D object generation in just 0.82 seconds on an A6000 GPU. Additionally, given the tokens enriched with hierarchical 3D-aware information, we finetune a pretrained LLM on them, enabling multimodal comprehension of 3D content.Our experiments show that SAR3D surpasses current 3D generation methods in both speed and quality and allows LLMs to interpret and caption 3D models comprehensively.
Human Mesh Recovery (HMR) from a single RGB image is a highly ambiguous problem, as an infinite set of 3D interpretations can explain the 2D observation equally well. Nevertheless, most HMR methods overlook this issue and make a single prediction without accounting for this ambiguity. A few approaches generate a distribution of human meshes, enabling the sampling of multiple predictions; however, none of them is competitive with the latest single-output model when making a single prediction. This work proposes a new approach based on masked generative modeling. By tokenizing the human pose and shape, we formulate the HMR task as generating a sequence of discrete tokens conditioned on an input image. We introduce MEGA, a MaskEd Generative Autoencoder trained to recover human meshes from images and partial human mesh token sequences. Given an image, our flexible generation scheme allows us to predict a single human mesh in deterministic mode or to generate multiple human meshes in stochastic mode. Experiments on in-the-wild benchmarks show that MEGA achieves state-of-the-art performance in deterministic and stochastic modes, outperforming single-output and multi-output approaches.
Styled Handwritten Text Generation (HTG) has recently received attention from the computer vision and document analysis communities, which have developed several solutions, either GAN- or diffusion-based, that achieved promising results. Nonetheless, these strategies fail to generalize to novel styles and have technical constraints, particularly in terms of maximum output length and training efficiency. To overcome these limitations, in this work, we propose a novel framework for text-image generation, dubbed Emuru. Our approach leverages a powerful text-image representation model (a variational autoencoder) combined with an autoregressive Transformer. Our approach enables the generation of styled text images conditioned on textual content and style examples, such as specific fonts or handwriting styles. We train our model solely on a diverse, synthetic dataset of English text rendered in over 100,000 typewritten and calligraphy fonts, which gives it the capability to reproduce unseen styles (both fonts and users' handwriting) in zero-shot. To the best of our knowledge, Emuru is the first autoregressive model for HTG, and the first designed specifically for generalization to novel styles. Moreover, our model generates images without background artifacts, which are easier to use for downstream applications. Extensive evaluation on both typewritten and handwritten, any-length text image generation scenarios demonstrates the effectiveness of our approach.
This paper presents the Large Vision Diffusion Transformer (LaVin-DiT), a scalable and unified foundation model designed to tackle over 20 computer vision tasks in a generative framework. Unlike existing large vision models directly adapted from natural language processing architectures, which rely on less efficient autoregressive techniques and disrupt spatial relationships essential for vision data, LaVin-DiT introduces key innovations to optimize generative performance for vision tasks. First, to address the high dimensionality of visual data, we incorporate a spatial-temporal variational autoencoder that encodes data into a continuous latent space. Second, for generative modeling, we develop a joint diffusion transformer that progressively produces vision outputs. Third, for unified multi-task training, in-context learning is implemented. Input-target pairs serve as task context, which guides the diffusion transformer to align outputs with specific tasks within the latent space. During inference, a task-specific context set and test data as queries allow LaVin-DiT to generalize across tasks without fine-tuning. Trained on extensive vision datasets, the model is scaled from 0.1B to 3.4B parameters, demonstrating substantial scalability and state-of-the-art performance across diverse vision tasks. This work introduces a novel pathway for large vision foundation models, underscoring the promising potential of diffusion transformers. The code and models are available at https://derrickwang005.github.io/LaVin-DiT.
State Space Models (SSMs) have shown significant promise in Natural Language Processing (NLP) and, more recently, computer vision. This paper introduces a new methodology leveraging Mamba and Masked Autoencoder (MAE) networks for point cloud data in both supervised and self-supervised learning. We propose three key contributions to enhance Mamba's capability in processing complex point cloud structures. First, we exploit the spectrum of a graph Laplacian to capture patch connectivity, defining an isometry-invariant traversal order that is robust to viewpoints and better captures shape manifolds than traditional 3D grid-based traversals. Second, we adapt segmentation via a recursive patch partitioning strategy informed by Laplacian spectral components, allowing finer integration and segment analysis. Third, we address token placement in MAE for Mamba by restoring tokens to their original positions, which preserves essential order and improves learning. Extensive experiments demonstrate our approach's improvements in classification, segmentation, and few-shot tasks over state-of-the-art (SOTA) baselines.
Multimodal entity alignment aims to identify equivalent entities in heterogeneous knowledge graphs by leveraging complementary information from multiple modalities. However, existing methods often overlook the quality of input modality embeddings during modality interaction – such as missing modality generation, modal information transfer, modality fusion – which may inadvertently amplify noise propagation while suppressing discriminative feature representations. To address these issues, we propose a novel model – CLAMEA for capturing latent modal association for multimodal entity alignment. Specifically, we use a self- attention mechanism to enhance salient information while attenuating noise within individual modality embeddings. We design a dynamic modal attention flow fusion module to capture and balance latent intra- and inter-modal associations and generate fused modality embeddings. Based on both fused and available modalities, we adopt variational autoencoder (VAE) to generate high quality embeddings for the missing modality. We use a cross-modal association extraction module to extract latent modal associations from the completed modality embeddings, further enhancing embedding quality. Experimental results on two real-world datasets demonstrate the effectiveness of our approach, which achieves an absolute 3.1% higher Hits@ 1 score than the sota method.