CVPR 2025 Papers

Immersive video offers a 6-Dof-free viewing experience, potentially playing a key role in future video technology. Recently, 4D Gaussian Splatting has gained attention as an effective approach for immersive video due to its high rendering efficiency and quality, though maintaining quality with manageable storage remains challenging. To address this, we introduce GIFStream, a novel 4D Gaussian representation using a canonical space and a deformation field enhanced with time-dependent feature streams. These feature streams enable complex motion modeling and allow efficient compression by leveraging their motion-awareness and temporal correspondence. Additionally, we incorporate both temporal and spatial compression networks for end-to-end compression. Experimental results show that GIFStream delivers high-quality immersive video at 30 Mbps, with real-time rendering and fast decoding on an RTX 4090.

Current video diffusion models achieve impressive generation quality but struggle in interactive applications due to bidirectional attention dependencies. The generation of a single frame requires the model to process the entire sequence, including the future. We address this limitation by adapting a pretrained bidirectional diffusion transformer to an autoregressive transformer that generates frames on-the-fly. To further reduce latency, we extend distribution matching distillation (DMD) to videos, distilling 50-step diffusion model into a 4-step generator. To enable stable and high-quality distillation, we introduce a student initialization scheme based on teacher's ODE trajectories, as well as an asymmetric distillation strategy that supervises a causal student model with a bidirectional teacher. This approach effectively mitigates error accumulation in autoregressive generation, allowing long-duration video synthesis despite training on short clips. Our model achieves a total score of 84.27 on the VBench-Long benchmark, surpassing all previous video generation models. It enables fast streaming generation of high-quality videos at 9.4 FPS on a single GPU thanks to KV caching. Our approach also enables streaming video-to-video translation, image-to-video, and dynamic prompting in a zero-shot manner. We release our code and pretrained models.

The objective of this study is to generate high-quality speech from silent talking face videos, a task also known as video-to-speech synthesis. A significant challenge in video-to-speech synthesis lies in the substantial modality gap between silent video and multi-faceted speech. In this paper, we propose a novel video-to-speech system that effectively bridges this modality gap, significantly enhancing the quality of synthesized speech. This is achieved by learning of hierarchical representations from video to speech. Specifically, we gradually transform silent video into acoustic feature spaces through three sequential stages -- content, timbre, and prosody modeling. In each stage, we align visual factors -- lip movements, face identity, and facial expressions -- with corresponding acoustic counterparts to ensure the seamless transformation. Additionally, to generate realistic and coherent speech from the visual representations, we employ a flow matching model that estimates direct trajectories from a simple prior distribution to the target speech distribution. Extensive experiments demonstrate that our method achieves exceptional generation quality comparable to real utterances, outperforming existing methods by a significant margin.

The rapid development of generative models has significantly advanced font generation. However, limited exploration has been devoted to the evaluation and interpretability of graphical fonts. Existing quality assessment models can only provide basic visual analyses, such as recognizing clarity and brightness, without in-depth explanations. To address these limitations, we first constructed a large-scale multimodal dataset named the Diversity Font Dataset (DFD), comprising 135,000 font-text pairs. This dataset encompasses a wide range of generated font types and annotations, including language descriptions and quality assessments, thus providing a robust foundation for training and evaluating font analysis models. Based on this dataset, we developed a font agent built upon a Vision-Language Model (VLM) aiming to enhance font quality assessment and offer interpretable question-answering capabilities. Alongside the original visual encoder in VLM, we integrated an Edge-Aware Traces (EAT) module to capture detailed edge information of font strokes and components. Furthermore, we introduced a Dynamic Direct Preference Optimization (D-DPO) strategy to facilitate efficient model fine-tuning. Experimental results demonstrate that Font-Agent achieves state-of-the-art performance on the established dataset. To further evaluate the generalization ability of our algorithm, we conducted additional experiments on several public datasets. The results highlight the notable advantage of Font-Agent in both assessing the quality of generated fonts and comprehending their content.

We study reconstructing and predicting 3D fluid appearance and velocity from a single video. Current methods require multi-view videos for fluid reconstruction. We present FluidNexus, a novel framework that bridges video generation and physics simulation to tackle this task. Our key insight is to synthesize multiple novel-view videos as references for reconstruction. FluidNexus consists of two key components: (1) a novel-view video synthesizer that combines frame-wise view synthesis with video diffusion refinement for generating realistic videos, and (2) a physics-integrated particle representation coupling differentiable simulation and rendering to simultaneously facilitate 3D fluid reconstruction and prediction. To evaluate our approach, we collect two new real-world fluid datasets featuring textured backgrounds and object interactions. Our method enables dynamic novel view synthesis, future prediction, and interaction simulation from a single fluid video. Project website: https://yuegao.me/FluidNexus.

We present FloVD, a novel video diffusion model for camera-controllable video generation. FloVD leverages optical flow to represent the motions of the camera and moving objects. This approach offers two key benefits. Since optical flow can be directly estimated from videos, our approach allows for the use of arbitrary training videos without ground-truth camera parameters. Moreover, as background optical flow encodes 3D correlation across different viewpoints, our method enables detailed camera control by leveraging the background motion. To synthesize natural object motion while supporting detailed camera control, our framework adopts a two-stage video synthesis pipeline consisting of optical flow generation and flow-conditioned video synthesis. Extensive experiments demonstrate the superiority of our method over previous approaches in terms of accurate camera control and natural object motion synthesis.

Video-language models have shown promise for addressing a range of multimodal tasks for video understanding, such as video question-answering. However, the inherent computational challenges of processing long video data and increasing model sizes have led to standard approaches that are limited by the number of frames they can process. In this work, we propose the Flexible Frame Selector (FFS), a learnable policy model with a new flexible selection operation, that helps alleviate input context restrictions by enabling video-language models to focus on the most informative frames for the downstream multimodal task, without adding undue processing cost. Our method differentiates from prior work due to its learnability, efficiency, and flexibility. We verify the efficacy of our method on standard video-question answering and reasoning benchmarks, and observe that our model can improve base video-language model accuracy while reducing the number of downstream processed frames.

We introduce Fish-Visual Trait Analysis (Fish-Vista), the first organismal image dataset designed for the analysis of visual traits of aquatic species directly from images using machine learning and computer vision methods. Fish-Vista contains 69,269 annotated images spanning 4,316 fish species, curated and organized to serve three downstream tasks: species classification, trait identification, and trait segmentation. Our work makes two key contributions. First, we provide a fully reproducible data processing pipeline to process fish images sourced from various museum collections, contributing to the advancement of AI in biodiversity science. We annotate the images with carefully curated labels from biological databases and manual annotations to create an AI-ready dataset of visual traits. Second, our work offers fertile grounds for researchers to develop novel methods for a variety of problems in computer vision such as handling long-tailed distributions, out-of-distribution generalization, learning with weak labels, explainable AI, and segmenting small objects. Dataset and code for Fish-Vista are available at https://github.com/Imageomics/Fish-Vista

The rapid growth of user-generated content (UGC) videos has produced an urgent need for effective video quality assessment (VQA) algorithms to monitor video quality and guide optimization and recommendation procedures. However, current VQA models generally only give an overall rating for a UGC video, which lacks fine-grained labels for serving video processing and recommendation applications. To address the challenges and promote the development of UGC videos, we establish the first large-scale Fine-grained Video quality assessment Database, termed FineVD, which comprises 6104 UGC videos with fine-grained quality scores and descriptions across multiple dimensions. Based on this database, we propose a Fine-grained Video Quality assessment (FineVQ) model to learn the fine-grained quality of UGC videos, with the capabilities of quality rating, quality scoring, and quality attribution. Extensive experimental results demonstrate that our proposed FineVQ can produce fine-grained video-quality results and achieve state-of-the-art performance on FineVD and other commonly used UGC-VQA datasets. Both FineVD and FineVQ are publicly available at: https://github.com/IntMeGroup/FineVQ.

Recent advancements in 2D and multimodal models have achieved remarkable success by leveraging large-scale training on extensive datasets. However, extending these achievements to enable free-form interactions and high-level semantic operations with complex 3D/4D scenes remains challenging. This difficulty stems from the limited availability of large-scale, annotated 3D/4D or multi-view datasets, which are crucial for generalizable vision and language tasks such as open-vocabulary and prompt-based segmentation, language-guided editing, and visual question answering (VQA). In this paper, we introduce Feature4X, a universal framework designed to extend any functionality from 2D vision foundation model into the 4D realm, using only monocular video input, which is widely available from user-generated content. The "X" in Feature4X represents its versatility, enabling any task through adaptable, model-conditioned 4D feature field distillation. At the core of our framework is a dynamic optimization strategy that unifies multiple model capabilities into a single representation. Additionally, to the best of our knowledge, Feature4X is the first method to distill and lift the features of video foundation models (e.g. SAM2, InternVideo2) into an explicit 4D feature field using Gaussian Splatting. Our experiments showcase novel view segment anything, geometric and appearance scene editing, and free-form VQA across all time steps, empowered by LLMs in feedback loops. These advancements broaden the scope of agentic AI applications by providing a foundation for scalable, contextually and spatiotemporally aware systems capable of immersive dynamic 4D scene interaction.

Face Forgery Video Detection (FFVD) is a critical yet challenging task in determining whether a digital facial video is authentic or forged. Existing FFVD methods typically focus on isolated spatial or coarsely fused spatiotemporal information, failing to leverage temporal forgery cues thus resulting in unsatisfactory performance. We strive to unravel these cues across three progressive levels: momentary anomaly, gradual inconsistency, and cumulative distortion. Accordingly, we design a consecutive correlate module to capture momentary anomaly cues by correlating interactions among consecutive frames. Then, we devise a future guide module to unravel inconsistency cues by iteratively aggregating historical anomaly cues and gradually propagating them into future frames. Finally, we introduce a historical review module that unravels distortion cues via momentum accumulation from future to historical frames. These three modules form our Temporal Forgery Cue Unraveling (TFCU) framework, sequentially highlighting spatial discriminative features by unraveling temporal forgery cues bidirectionally between historical and future frames. Extensive experiments and ablation studies demonstrate the effectiveness of our TFCU method, achieving state-of-the-art performance across diverse unseen datasets and manipulation methods. Code is available at https://github.com/zhenglab/TFCU.

Progress in machine understanding of sign languages has been slow and hampered by limited data. In this paper, we present FSboard, an American Sign Language fingerspelling dataset situated in a mobile text entry use case, collected from 147 paid and consenting Deaf signers using Pixel 4A selfie cameras in a variety of environments. Fingerspelling recognition is an incomplete solution that is only one small part of sign language translation, but it could provide some immediate benefit to Deaf/Hard of Hearing signers while more broadly capable technology develops. At >3 million characters in length and >250 hours in duration, FSboard is the largest fingerspelling recognition dataset to date by a factor of >10x. As a simple baseline, we finetune 30 Hz MediaPipe Holistic landmark inputs into ByT5-Small and achieve 11.1% Character Error Rate (CER) on a test set with unique phrases and signers. This quality degrades gracefully when decreasing frame rate and excluding face/body landmarks---plausible optimizations to help with on-device performance---but falls short of human performance measured at 2.2% CER.

Figure skating, known as the "Art on Ice," is among the most artistic sports, challenging to understand due to its blend of technical elements (like jumps and spins) and overall artistic expression. Existing figure skating datasets mainly focus on single tasks, such as action recognition or scoring, lacking comprehensive annotations for both technical and artistic evaluation. Current sports research is largely centered on ball games, with limited relevance to artistic sports like figure skating. To address this, we introduce FSAnno, a large-scale dataset advancing artistic sports understanding through figure skating. FSAnno includes an open-access training and test dataset, alongside a benchmark dataset, FSBench, for fair model evaluation. FSBench consists of FSBench-Text, with multiple-choice questions and explanations, and FSBench-Motion, containing multimodal data and Question and Answer (QA) pairs, supporting tasks from technical analysis to performance commentary. Initial tests on FSBench reveal significant limitations in existing models' understanding of artistic sports. We hope FSBench will become a key tool for evaluating and enhancing model comprehension of figure skating. All data, models, and more details are available at: https://github.com/Moomin-Fin/Ano.

Egocentric motion capture with a head-mounted body-facing stereo camera is crucial for VR and AR applications but presents significant challenges such as heavy occlusions and limited annotated real-world data. Existing methods rely on synthetic pretraining and struggle to generate smooth and accurate predictions in real-world settings, particularly for lower limbs. Our work addresses these limitations by introducing a lightweight VR-based data collection setup with on-board, real-time 6D pose tracking. Using this setup, we collected the most extensive real-world dataset for ego-facing ego-mounted cameras to date in size and motion variability. Effectively integrating this multimodal input -- device pose and camera feeds -- is challenging due to the differing characteristics of each data source. To address this, we propose FRAME, a simple yet effective architecture that combines device pose and camera feeds for state-of-the-art body pose prediction through geometrically sound multimodal integration and can run at 300 FPS on modern hardware. Lastly, we showcase a novel training strategy to enhance the model's generalization capabilities. Our approach exploits the problem's geometric properties, yielding high-quality motion capture free from common artifacts in prior works. Qualitative and quantitative evaluations, along with extensive comparisons, demonstrate the effectiveness of our method. Data, code, and CAD designs will be available at https://vcai.mpi-inf.mpg.de/projects/FRAME/

Learned Video Compression (LVC) aims to reduce redundancy in sequential data through deep learning approaches. Recent advances have significantly boosted LVC performance by shifting compression operations to feature domain, often combining Motion Estimation and Motion Compensation module(MEMC) with CNN-based context extraction. However, reliance on motions and convolution-driven context models limits generalizability and global perception. To address these issues, we propose a Feature-level Attention (FLA) module within a Transformer-based framework that perceives full-frame explicitly, thus bypassing confined motion signatures. FLA accomplishes global perception by converting high-level local patch embeddings to one-dimensional batch-wise vectors and replacing traditional attention weights to a global context matrix. Amongst this, a dense overlapping patcher (DP) is introduced to retain local features before embedding projection. Furthermore, a Transformer-CNN mixed encoder is applied to alleviate the spatial feature bottleneck without expanding latent size. Experiments demonstrate excellent generalizability with universally efficient redundancy reduction in different scenarios. Extensive tests on four video compression datasets show that our method achieves state-of-the-art Rate-Distortion performance compared to existing LVC methods and traditional codecs. A down-scaled version of our model reduced computation overhead by a great margin while maintained great performance.

Anticipating how a person will interact with objects in an environment is essential for activity understanding, but existing methods are limited to the 2D space of video frames--capturing physically ungrounded predictions of "what" and ignoring the "where" and "how". We introduce FIction for 4D future interaction prediction from videos. Given an input video of a human activity, the goal is to predict which objects at what 3D locations the person will interact with in the next time period (e.g., cabinet, fridge), and how they will execute that interaction (e.g., poses for bending, reaching, pulling). Our novel model FIction fuses the past video observation of the person's actions and their environment to predict both the "where" and "how" of future interactions. Through comprehensive experiments on a variety of activities and real-world environments in Ego-Exo4D, we show that our proposed approach outperforms prior autoregressive and (lifted) 2D video models substantially, with more than 30% relative gains.

Gaze direction serves as a pivotal indicator for assessing the level of driver attention. While image-based gaze estimation has been extensively researched, there has been a recent shift towards capturing gaze direction from video sequences. This approach encounters notable challenges, including the comprehension of the dynamic pupil evolution across frames and the extraction of head pose information from a relatively static background. To surmount these challenges, we introduce a dual-stream deep learning framework that explicitly models the displacement changes of the pupil through a fine-grained inter-frame attention mechanism and generates weights to adjust gaze embeddings. This technique transforms the face into a set of distinct patches and employs cross-attention to ascertain the correlation between pixel displacements in various patches and adjacent frames, thereby tracking spatial dynamics within the sequence. Our method is validated using two publicly available driver gaze datasets, and the results indicate that it achieves state-of-the-art performance or is on par with the best outcomes while reducing the parameters.

Reconstructing high-fidelity, animatable 3D head avatars from effortlessly captured monocular videos is a pivotal yet formidable challenge. Although significant progress has been made in rendering performance and manipulation capabilities, notable challenges remain, including incomplete reconstruction and inefficient Gaussian representation. To address these challenges, we introduce FATE -- a novel method for reconstructing an editable full-head avatar from a single monocular video. FATE integrates a sampling-based densification strategy to ensure optimal positional distribution of points, improving rendering efficiency. A neural baking technique is introduced to convert discrete Gaussian representations into continuous attribute maps, facilitating intuitive appearance editing. Furthermore, we propose a universal completion framework to recover non-frontal appearance, culminating in a 360^\circ-renderable 3D head avatar. FATE outperforms previous approaches in both qualitative and quantitative evaluations, achieving state-of-the-art performance. To the best of our knowledge, FATE is the first animatable and 360^\circ full-head monocular reconstruction method for a 3D head avatar. Project page and code are available at this \href https://zjwfufu.github.io/FATE-page/ link .

Continual Learning in semantic scene segmentation aims to continually learn new unseen classes in dynamic environments while maintaining previously learned knowledge. Prior studies focused on modeling the catastrophic forgetting and background shift challenges in continual learning. However, fairness, another major challenge that causes unfair predictions leading to low performance among major and minor classes, still needs to be well addressed. In addition, prior methods have yet to model the unknown classes well, thus resulting in producing non-discriminative features among unknown classes. This work presents a novel Fairness Learning via Contrastive Attention Approach to continual learning in semantic scene understanding. In particular, we first introduce a new Fairness Contrastive Clustering loss to address the problems of catastrophic forgetting and fairness. Then, we propose an attention-based visual grammar approach to effectively model the background shift problem and unknown classes, producing better feature representations for different unknown classes. Through our experiments, our proposed approach achieves State-of-the-Art (SoTA) performance on different continual learning benchmarks, i.e., ADE20K, Cityscapes, and Pascal VOC. It promotes the fairness of the continual semantic segmentation model.