paperlist.ai

AI conference papers

AI conference paper library

Search papers from 30+ top conferences, including NeurIPS, ICLR, ICML, CVPR, ACL and AAAI. Filter by keyword, then open any paper for its abstract, authors and links.

ECCV 2026

SOMA: From Surface Observations to Muscle Anatomy

Eduardo Alvarado, Emily Kim, Friedemann Runte, Gerrit Nolte, Mario Botsch, Marc Habermann, Christian Theobalt

With the growing demand for realistic virtual humans, para-metric body models have become a cornerstone of modern medicine, sportsor entertainment applications. However, most of these models are inherentlylimited: they only capture the 3D surface of the skin, offering no insightinto the complex bio-mechanical structures that generate motion. As moreapplications expand towards biomechanics, the need for virtual humanmodels that go beyond the skin has become increasingly evident. Traditionalsoft-tissue simulations, such as FEM, are accurate but non-scalable and toocomputationally expensive for most common applications. Alternatively,existing biomechanical tools can simulate muscular forces and activations,but do not model changes in external shape, restricting how activationscorrelate with actual observable anatomy. This motivates a novel inverseresearch problem: recovering muscle deformations directly from visiblesurface observations - i.e., from the skin, and thus the pose. In this work,we present SOMA (from Surface Observations to Muscle Anatomy), aperson-specific model that infers spatio-temporal muscle behavior fromsurface signals obtained using RGB cameras, and SKIM, a subject-specificsoft-tissue deformation dataset. To the best of our knowledge, this is thefirst method that attempts to recover muscle deformations from multi-viewRGB data. We show how our method provides anatomically groundedanimations without the complexity of traditional simulations, leading to ascalable and cost-effective solution. Data and code are available.

ECCV 2026

SIGNER: Temporally Grounded Sign Language Generation via Time-Resolved Conditioning

Taeryung Lee, Hyeongjin Nam, Gyeongsik Moon, Kyoung Mu Lee

Sign language generation (SLG), also known as text-to-signgeneration, aims to bridge the communication gap between signers andnon-signers. Unlike many other generative tasks, SLG must satisfy twofundamental linguistic constraints. First, sign language expresses mean-ing through a sequence of gestures aligned with word-like units calledglosses, and therefore requires correct lexical ordering to preserve in-tended meaning. Second, each gesture should faithfully reflect the in-tended gloss (semantic accuracy). Despite recent progress, existing SLGmethods frequently produce signs with incorrect lexical order and lowsemantic accuracy. A common limitation of prior approaches stems fromglobally fused conditioning strategies, which weaken temporal ground-ing, the temporal correspondence between glosses and their realized signsegments. This often leads to incorrect lexical order and semanticallyambiguous signs. To address this limitation, we propose SIGNER, aSIGN language generation framework with timE-Resolved conditioningto ensure temporal grounding, leveraging a temporal-gloss condition andlocal temporal fusion (LTF). SIGNER constructs a temporal-gloss con-dition by estimating a gloss sequence and its durations from input text,and assigning gloss semantics across the temporal dimension. We thenintroduce LTF, a temporally grounded fusion module that integratesthe temporal-gloss condition within a constrained temporal window dur-ing denoising. By enforcing temporal locality in condition fusion, LTFpreserves temporal grounding, leading to correct lexical ordering andclearer per-gloss semantics. Experiments on Phoenix-2014T and CSL-Daily demonstrate state-of-the-art performance, further supported bymotion-smoothness analysis. The project page is available here.

ECCV 2026

HEM: a margin-based loss for visual categorisation tasks

Michael Spratling, Heiko Schütt

Training deep neural networks (DNNs) on classi(cid:28)cation tasks can be performed with a number of di(cid:27)erent losses, but cross-entropy (CE) loss is the de-facto standard. Here, we propose an alternative loss, high error margin (HEM), which is a margin based loss modi(cid:28)ed to improve the training dynamics of neural networks. HEM loss is evaluated extensively using a wide range of DNN architectures and benchmark datasets with all experimental settings and training hyper-parameters taken from the literature, and hence, optimised for CE loss. HEM is found to be more e(cid:27)ective than CE loss across a range of image-based tasks: unknown class rejection, adversarial robustness, learning with imbalanced data, continual learning, and semantic segmentation (a pixel-wise classi- (cid:28)cation task). HEM is inferior to CE only in terms of clean and corrupt image classi(cid:28)cation with balanced training data, and this di(cid:27)erence is small. We also compare HEM to specialised losses that have previously been proposed to improve performance for speci(cid:28)c vision tasks. LogitNorm, a loss achieving state-of-the-art performance on unknown class rejection, produces similar performance to HEM for this task, but is much poorer for continual learning and semantic segmentation. Logit-adjusted loss, designed for imbalanced data, has superior results to HEM for that task, but performs worse on unknown class rejection and semantic segmentation. DICE, a popular loss for semantic segmentation, is inferior to HEM loss on all tasks, including semantic segmentation. Overall, HEM is competitive with the best alternative loss for all the tasks we have used and performs better than all other tested losses in terms of rejecting out-of-distribution examples, for continual learning, and by a substan- 1 tial margin for semantic segmentation . • • •

ECCV 2026

Segmenting, Fast and Slow: Real-Time Open-Vocabulary Video Instance Segmentation with Dual-Path Processing

Luca Barsellotti, Martin Sundermeyer, Mattia Segu, Nikita Araslanov, Muhammad Ferjad Naeem, Marcella Cornia, Yongqin Xian, Maxim Berman

Object-centric models inspired by DETR have become thedominant paradigm for open-vocabulary video instance segmentation(OV-VIS). While recent efforts have reduced the computational cost ofpixel decoding, textual modality fusion, and object decoding to makethese architectures more suitable for mobile devices, real-time on-deviceinference at high frame rates remains an open challenge. In this paper,we introduce SegFS, a dual-stream fast-slow framework that significantlyimproves efficiency without sacrificing accuracy. On sparse keyframes, anopen-vocabulary object-based model predicts instance-level representa-tions. These representations are then projected back into the backbonefeature space to condition a lightweight fast network, which efficientlyrelocalizes and segments the instances in subsequent frames. By shiftinginstance propagation from object decoding to feature-space condition-ing, our approach decouples multimodal semantic understanding fromdense mask prediction and enables efficient temporal propagation. Theproposed fast branch achieves up to 14× lower latency than the mobile-oriented MOBIUS model, while maintaining competitive segmentationperformance on standard OV-VIS benchmarks.

ECCV 2026

SARIF: Segment Anything for Robust Image Forensics

Dong-Hyun Moon, Ju-Hyeon Nam, Sang-Chul Lee

Image forgery localization remains challenging due to di-verse manipulation techniques and distribution shifts. Existing recentforgery localization models achieve high accuracy on benchmarks butoften struggle with cross-domain generalization and robustness. In thispaper, we propose SARIF (Segment Anything for Robust Im-age Forensics), a framework that leverages Segment Anything Model(SAM), which has a promptable architecture and generalization abilityto overcome these limitations. SARIF introduces a feedback-guided maskdecoder and a dual-encoder design that extracts forgery-specific informa-tion to capture forensic traces while exploiting SAM’s architecture. Tolocalize manipulated regions, we design a block-wise prompting mecha-nism that derives forgery-specific cues from residual features between anadapted encoder and its frozen counterpart. These features are fused withthe previous mask prompt to drive a feedback-based mask refinementprocess, enabling automatic forgery segmentation without manual in-put. Extensive experiments on standard forgery-localization benchmarksshow that SARIF achieves strong average cross-dataset performance androbustness to common image corruptions. Our SARIF code is availablein GitHub Link.

ECCV 2026

Robobench: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models as Embodied Brain

Yulin Luo, Chun-Kai Fan, Menghang Dong, Jiayu Shi, Xiangju Mi, Mengdi Zhao, Bo-Wen Zhang, Cheng Chi, Jiaming Liu, Gaole Dai, Rongyu Zhang, Ruichuan An, Kun Wu, Zhengping Che, Shaoxuan Xie, Guocai Yao, Zhongxia Zhao, Pengwei Wang, Guang Liu, Zhongyuan Wang, Tiejun Huang, Shanghang Zhang

Building robots that can perceive, reason, and act in dy-namic, unstructured environments remains a core challenge. Recent em-bodied systems often adopt a dual-system paradigm, where System 2handles high-level reasoning while System 1 executes low-level control.In this work, we refer to System 2 as the embodied brain, emphasizingits role as the cognitive core for reasoning and decision-making in ma-nipulation tasks. Given this role, systematic evaluation of the embodiedbrain is essential for advancing robotic intelligence. Yet existing bench-marks emphasize execution success, or, when targeting high-level reason-ing, su_x001B_er from incomplete dimensions and limited task realism, o_x001B_eringonly a partial picture of cognitive capability. To bridge this gap, we intro-duce RoboBench, a benchmark that systematically evaluates multimodallarge language models (MLLMs) as embodied brains. Motivated by thedistinct cognitive roles required across the full manipulation pipeline,RoboBench de_x001C_nes _x001C_ve dimensions_x0016_Instruction Comprehension, Percep-tion Reasoning, Generalized Planning, A_x001B_ordance Prediction, and Fail-ure Analysis_x0016_spanning 14 capabilities, 25 tasks, and 6092 QA pairs. Toensure realism, we curate datasets across diverse embodiments, attribute-rich objects, multi-view scenes, and memory-driven navigation, drawingfrom large-scale real robotic data and in-house collection. For planning,RoboBench introduces an evaluation framework that uses an MLLM as aworld simulator. It moves beyond symbolic matching to evaluate embod-ied feasibility by simulating whether predicted plans can achieve criticalobject-state changes under physical and visual constraints, enabling faith-ful assessment of long-horizon reasoning. Experiments on 18 state-of-the-art MLLMs reveal fundamental limitations: di_x001E_culties with implicit in-struction comprehension, spatiotemporal reasoning, cross-scenario plan-ning, _x001C_ne-grained a_x001B_ordance understanding, and execution failure diagno-Dimension Subdimension Performance DownstreamInstuctionComprehension Explicit Object Goal Implicit Demand GoalPerception Robotic- Object- Scene- Task-Reasoning Centric Centric Centric CentricGeneralization EmbodimentCross- Cross- Cross- Cross-Planning Object View TaskAffordance Static Dynamic NavigationPrediction Affordance Affordance AffordanceFailure Low-Level High-LevelAnanlysis Execution Error Planning Error. Instuction Comprehension - . Long-Horizon Planning - . Next-Step Planning - . Static AffordanceImplicit Demand Goal: Question: Question: Question:Please help me tidy up the apple How to complete the goal? What's the next subgoal? What's the contact point?Put the apple into the drawer open(drawer) pick_up(apple) open(drawer) [x , y ]place(apple, drawer) close(drawer)decomposelong-horizen goal adapt subgoalinto substeps in closed-loopinterpret gather diagnose, monitor state refine subgoal intohuman task-relevant analyze, in closed-loop Yes spatial aligned cuesintention. Perception Reasoning adjust error - . Dynamic Affordanceinformation . Failure Analysis - . Task State EstimationTask-Related:robotic type? single arm Question: Question: Question:object attribute? apple: rigid Is there any error? What type of error? Has current subgoal been completed? What's the movement trajectory?operation note? drawer: closed Yes, high-level error---missing steps Yes No [x , y ], [x ,y ], [x , y ],[x , y ]spatial relation? apple lower than drawermonitor state refine subgoal intoin closed-loop spatial aligned cuesNoFig. 1: Overview of RoboBench We evaluate MLLMs as embodied brains with25 tasks color-coded by 5 dimensions (top left). These dimensions follow the em-bodied execution pipeline (bottom)_x0016_from intent understanding to failure diagnos-ing_x0016_capturing the core cognitive roles of System 2. Performance comparison (top mid-dle) reveals signi_x001C_cant gaps among state-of-the-art MLLMs (top right) RoboBenchscores strongly correlate with downstream VLA performance on CALVIN.sis. We further analyze how embodied cognitive abilities relate to down-stream robotic control. RoboBench provides a comprehensive sca_x001B_oldto quantify high-level cognition, clarify the role of the embodied brain,and guide the development of next-generation MLLMs for more robustrobotic intelligence. Project page: https://robo-bench.github.io.

ECCV 2026

Rethinking Garment Conditioning in Diffusion-based Virtual Try-On: Decouple, Don't Denoise

Kihyun Na, Jinyoung Choi, Injung Kim

Virtual Try-On (VTON) synthesizes realistic images of aperson wearing a target garment, with broad applications in e-commerceand fashion. Diffusion-based dual-UNet methods achieve strong resultsbut double the parameters by dedicating a separate network to garmentconditioning. Spatial concatenation offers a simpler single-network alter-native, yet both UNet- and DiT-based instantiations report that full fine-tuning is ineffective, and the community has settled for attention-onlytraining. We ask: why does full fine-tuning fail, and can this be resolved?Through what is, to our knowledge, the first visualization study of dual-UNet reference network behavior, we identify a unifying insight: garmentconditioning must be decoupled from the denoising process. Spatial con-catenation violates this by embedding the garment within the denoisingtarget, causing three conflicts: guidance leakage, gradient competition,and train-test discrepancy. We derive three design principles to restorethis decoupling and implement them as a pure recipe atop a standardarchitecture with no modification. The resulting model, DeCo-VTON(860M params), achieves single-network state of the art, matching thedual-UNet state of the art at half the cost while being preferred in humanevaluation.

ECCV 2026

F⁴Splat: Feed-Forward Predictive Densification for Feed-Forward 3D Gaussian Splatting

Injae Kim, Chaehyeon Kim, Minseong Bae, Minseok Joo, Hyunwoo Kim

Feed-forward 3D Gaussian Splatting methods enable single-pass reconstruction and real-time rendering. However, they typicallyadopt rigid pixel-to-Gaussian or voxel-to-Gaussian pipelines that uni-formly allocate Gaussians, leading to redundant Gaussians across views.Moreover, they lack an effective mechanism to control the total numberof Gaussians while maintaining reconstruction fidelity. To address theselimitations, we present F4 Splat, which performs Feed-Forward predic-tive densification for Feed-Forward 3D Gaussian Splatting, introduc-ing a densification-score-guided allocation strategy that adaptively dis-tributes Gaussians according to spatial complexity and multi-view over-lap. Our model predicts per-region densification scores to estimate therequired Gaussian density and allows explicit control over the final Gaus-sian budget without retraining. This spatially adaptive allocation reducesredundancy in simple regions and minimizes duplicate Gaussians acrossoverlapping views, producing compact yet high-quality 3D representa-tions. Extensive experiments demonstrate that our model achieves su-perior novel-view synthesis performance compared to prior uncalibratedfeed-forward methods, while using significantly fewer Gaussians.

ECCV 2026

RelAfford6D: Relational 6D Affordance Graphs for Constraint-Driven Robotic Manipulation

Guodong Zhang, Qichen He, Wenyuan XIE, Shaokai Wu, Yanbiao Ji, Qiuchang Li, Bayram Bayramli, Yue Ding, Hongtao Lu

Bridging abstract semantics and precise physical control re-mains a fundamental challenge in open-world robotic manipulation. Whilerecent data-driven policies show promise, their reliance on isolated con-tact points or latent affordance embeddings lacks the rigorous kinematicconstraints necessary for complex articulated objects.To overcome thelimitation, we introduce RelAfford6D, a novel training-free frameworkcentered on a Relational 6D Affordance Graph. Given a free-form in-struction, our system deduces a semantic topology linking a primary in-teracting part to its physical anchor. By elevating these topological nodesinto precise metric SE(3) poses via vision foundation models, we analyti-cally formulate downstream execution as a kinematic constraint satisfac-tion problem. The robot synthesizes continuous trajectories by trackingstrictly defined physical manifolds (e.g., revolute or prismatic orbits).Coupled with a closed-loop tracking mechanism for dynamic replanningagainst disturbances, our physically grounded approach achieves supe-rior zero-shot success rates, cross-category generalization and executionrobustness in both simulation and the real world environments, outper-forming existing data-driven baselines.

ECCV 2026

ReconPhys: Reconstruct Appearance and Physical Attributes from Single Video

Boyuan Wang, Xiaofeng Wang, Yongkang Li, Zheng Zhu, Yifan Chang, Angen Ye, Guosheng Zhao, Chaojun Ni, Guan Huang, Yijie Ren, Yueqi Duan, Xingang Wang

Reconstructing non-rigid objects with physical plausibilityremains a significant challenge. Existing approaches leverage differen-tiable rendering for per-scene optimization, recovering geometry anddynamics but requiring expensive tuning or manual annotation, whichlimits practicality and generalizability. To address this, we propose Re-conPhys, the first feedforward framework that jointly learns physicalattribute estimation and 3D Gaussian Splatting reconstruction from asingle monocular video. Our method employs a dual-branch architecturetrained via a self-supervised strategy, eliminating the need for ground-truth physics labels. Given a video sequence, ReconPhys simultaneouslyinfers geometry, appearance, and physical attributes. Experiments ona large-scale synthetic dataset demonstrate superior performance: ourmethod achieves 21.64 PSNR in future prediction compared to 13.27by state-of-the-art optimization baselines, while reducing Chamfer Dis-tance from 0.349 to 0.004. Crucially, ReconPhys enables fast inference(<1 second) versus hours required by existing methods, facilitating rapidgeneration of simulation-ready assets for robotics and graphics. The codeis available at https://github.com/chuanshuogushi/ReconPhys.

ECCV 2026

LivingWorld: Interactive 4D World Generation with Environmental Dynamics

Hyeongju Mun, In-Hwan Jin, Sohyeong Kim, Kyeongbo Kong

We introduce LivingWorld, an interactive framework forgenerating 4D worlds with environmental dynamics from a single image.While recent advances in 3D scene generation enable large-scale environ-ment creation, most approaches focus primarily on reconstructing staticgeometry, leaving scene-scale environmental dynamics such as clouds,water, or smoke largely unexplored. Modeling such dynamics is challeng-ing because motion must remain coherent across an expanding scenewhile supporting low-latency user feedback. LivingWorld addresses thischallenge by progressively constructing a globally coherent motion field asthe scene expands. To maintain global consistency during expansion, weintroduce a geometry-aware alignment module that resolves directionaland scale ambiguities across views. We further represent motion using acompact hash-based motion field, enabling efficient querying and stablepropagation of dynamics throughout the scene. This representation alsosupports bidirectional motion propagation during rendering, producinglong and temporally coherent 4D sequences without relying on expen-sive video-based refinement. On a single RTX 5090 GPU, generatingeach new scene expansion step requires 9 seconds, followed by 3 secondsfor motion alignment and motion field updates, enabling interactive 4Dworld generation with globally coherent environmental dynamics. Videodemonstrations are available at paper.pnu-cvsp.com/LivingWorld.

ECCV 2026

Prompt2Effect: Training-Free LoRA Synthesis for Controllable Video Effects

Xiaomeng Yang, Yanyu Li, Gordon Qian, Ivan Skorokhodov, Viacheslav Ivanov, Avalon Vinella, Xuan Zhang, Yanzhi Wang, Sergey Tulyakov, Anil Kag

While personalizing Image-to-Video (I2V) diffusion modelswith specific visual effects is increasingly demanded for high-end genera-tion, current practice requires training a separate Low-Rank Adaptation(LoRA) module for each effect, incurring substantial data curation anditerative optimization costs that hinder interactive control. We presentPrompt2Effect, a weight-driven hypernetwork that amortizes per-effecttraining by directly synthesizing effect-specific LoRA weights in a singleforward pass. Unlike prior hypernetworks that regress adapter weightspurely from semantics, Prompt2Effect is explicitly conditioned on thefrozen base model weights, grounding prediction in the structural geom-etry of each layer. Furthermore, instead of predicting raw LoRA matri-ces, we introduce an SVD-canonicalized parameterization that resolvesfactorization ambiguity and stabilizes large-scale synthesis. Extensive ex-periments demonstrate that Prompt2Effect achieves on-par or superiorvideo quality and effect alignment compared to conventional LoRA fine-tuning, while reducing the computational cost from 56 GPU traininghours to 3.3 seconds of hypernetwork inference. When used as initializa-tion for subsequent fine-tuning, our predicted weights further improvefinal performance and accelerate optimization by approximately 10×.

ECCV 2026

PointLAM: Local Attentive Mamba for Efficient Point-based 3D Object Detection

Xuanming Shang, Weijia Zhang, Chao Ma

3D object detection from LiDAR point clouds faces a fun-damental dilemma: voxel-based methods achieve efficiency at the costof geometric quantization, while point-based methods preserve fidelitybut suffer from prohibitive computational bottlenecks. Specifically, point-based architectures are crippled by slow downsampling strategies (e.g.,FPS) and expensive dynamic neighbor queries (e.g., k-NN) coupled withcostly continuous interactions. To tackle these systemic inefficiencies, wepropose PointLAM, a highly efficient and powerful point-based architec-ture driven by two synergistic innovations. First, to resolve the downsam-pling bottleneck, we develop the Laplacian Point Sampler (LPS). LPSemploys an implicit discrete Laplacian high-pass filter and Doubly SortedSampling to achieve fast, structure-aware foreground preservation. Sec-ond, to overcome local modeling latency, we design the Local HadamardAggregator (LHA). LHA decouples spatial indexing from feature repre-sentation using transient grids, and replaces complex continuous interac-tions with a Hadamard Gating mechanism for topology-aware, attentivemodulation. By coupling this local gating with Bi-Directional Mamba(BDM) layers for global sequence modeling, we formulate the Local At-tentive Mamba (LAM) block. Powered by this architecture, PointLAMachieves competitive performance on nuScenes and Waymo for point-based detectors. It rivals highly optimized voxel competitors while re-quiring a fraction of the computational footprint, demonstrating markedsuperiority in detecting small instances and handling extreme sparsity.Project page: https://pointlam.github.io/.

ECCV 2026

SplatPainter: Interactive Authoring of 3D Gaussians from 2D Edits via Test-Time Training

Yang Zheng, Hao Tan, Kai Zhang, Peng Wang, Leonidas Guibas, Gordon Wetzstein, Wang Yifan

The rise of 3D Gaussian Splatting has revolutionized pho-torealistic 3D asset creation, yet a critical gap remains for their inter-active appearance refinement and editing. Existing approaches based ondiffusion or optimization are ill-suited for this task, as they are oftenprohibitively slow, lack the precision for fine-grained control, or—mostimportantly—are fundamentally destructive to the original asset’s iden-tity. To address this, we introduce SplatPainter, a state-aware feedfor-ward model that enables continuous, high-fidelity appearance editing of3D Gaussian assets from user-provided 2D views. Our method directlypredicts updates to the attributes of a compact, feature-rich Gaussianrepresentation and leverages Test-Time Training to create a state-aware,iterative workflow. The versatility of our approach allows a single ar-chitecture to perform diverse tasks, including high-frequency local detailrefinement, local paint-over, and consistent global recoloring, all at inter-active speeds while strictly preserving the asset’s original structure. Ourproject website is at: https://y-zheng18.github.io/SplatPainter/.

ECCV 2026

PixelPilot: Scalable Vision-Language-Action Models for End-to-End Autonomous Driving

Pin Tang, Guoqing Wang, Xiangxuan Ren, Zhongdao Wang, Guodongfang Zhao, Bailan Feng, Chao Ma

Vision-Language-Action Models (VLAs), which leverage theadvanced reasoning capabilities of Vision-Language Models (VLMs), showpromising generalization in complex autonomous driving scenarios. Ex-isting VLAs typically predict and optimize 3D trajectories from 2D im-ages. While intuitive, this 2D-to-3D prediction is inherently entangledwith camera parameters, leading to limited data scalability across het-erogeneous driving datasets. Moreover, directly optimizing in 3D spaceinduces severe convergence to trivial solutions, where VLAs rely on ego-status rather than visual scene understanding. To address these issues,we propose PixelPilot, a novel VLA featuring a decoupled planning andlifting paradigm. In the planning phase, PixelPilot reformulates scene un-derstanding and trajectory prediction as sensor-agnostic 2D-to-2D tasksin the image plane, thereby facilitating scalable training across diversedatasets. The planned 2D trajectories are then deterministically lifted to3D only during inference, ensuring the full exploitation of visual cues andgeneralization across different vehicles. To realize this paradigm, we pro-pose a knowledge-instilled policy learning strategy that applies dense, in-termediate rewards via Group Relative Policy Optimization (GRPO) toenforce a rigorous causal chain from visual perception to spatial planning.Extensive experiments demonstrate that PixelPilot achieves state-of-the-art performance in both open-loop and closed-loop settings, validatingits superior scalability and visual reasoning capabilities.

ECCV 2026

PhysAlign: Learning Physical Priors for Dynamical Event-Driven Video Generation via Representation Alignment

Mengxian Li, Zhan Wang, Fan Qi, Changsheng Xu

Current video generation models produce highly realistic vi-suals but frequently fail to maintain physical and causal consistencywhen facing complex dynamic events, such as collisions or collapses.We attribute this limitation to the fact that existing models primar-ily fit visual data distributions, entangling dynamics with appearancewithout explicit physical priors to govern motion evolution. To addressthis, we introduce PhysAlign, a framework for Dynamical Event-DrivenPhysically Consistent Video Generation. PhysAlign decouples physicalpriors from visual appearance via a Physical Dynamics Extractor anda discrete Universal Physical Codebook. Recognizing the temporal lo-cality of dynamic events, an Event-Aware Temporal Gating module dy-namically controls the injection of physical priors, while Physics-GuidedAlignment distills continuous physical topologies from a video founda-tion model. Furthermore, we construct EventPhy, a 26K-video datasetwith structured, VLM-generated causal annotations to benchmark dy-namical event-driven video generation. Extensive evaluations show thatPhysAlign achieves state-of-the-art physical reasoning, obtaining leadingSemantic Adherence and Physical Commonsense scores on challengingbenchmarks like PhysicsIQ and VideoPhy, while preserving high visualfidelity. Project Page: https://github.com/FanQi-AI/PhysAlign

ECCV 2026

From smooth to sharp: Frequency-Decoupled Latent Optimization for Realistic Image Generation

Tejaswini Medi, Hsien-Yi Wang, Arianna Rampini, Margret Keuper

Latent generative models compress images into learned em-beddings prior to synthesis, and the generation quality critically dependson how faithfully these embeddings preserve visual detail. We observethat while such embeddings are effective at reconstructing low frequencystructure, they struggle to recover sharp high frequency details that areessential for perceptual realism. Conventional reconstruction objectivesimplicitly prioritize coarse structural information over high frequencycontent, which can lead to overly smoothed outputs and degraded vi-sual quality in textured regions. Motivated by this observation, we pro-pose DeBaT, a Decoupled frequency Band Tokenizer that explicitly sep-arates the learning of low and high frequency band embeddings. Thisdecoupling enables accurate reconstruction of fine details while preserv-ing global coherence. Integrated into a latent diffusion based generativemodel, DeBaT allows for sharper and more realistic samples than previ-ous latent tokenizers, confirming that the explicit decoupling of high andlow frequency bands eases the preservation of visual details in learnedembedding spaces.

ECCV 2026

Parallax Portrait Matting

Xin Cai, Jiawen Chen, Lars Jebe, Tianfan Xue, Zhoutong Zhang

Image matting is highly ill-posed, especially when both theforeground and background are richly textured. While single-image mat-ting methods learn strong priors from data, they often struggle on thesechallenging cases. Existing approaches improve results by requiring ad-ditional signals such as green screens, polarized lighting, or clean back-ground images, but these typically rely on specialized capture setups.We present Parallax Portrait Matting, a practical two-frame mattingmethod that uses a second image captured with slight viewpoint change.Such a setting arises naturally in burst photography, where small cameramotion induces foreground-background parallax and provides comple-mentary observations for matting. Our pipeline estimates trimaps andforeground/background motion, then constructs aligned views for pre-diction. To handle imperfect motion estimation, the network uses thebackground-aligned pair for direct fusion and the foreground-aligned cuethrough cross-attention for error compensation. Experiments show thatour method recovers finer details and more accurate foreground colorsthan strong single-image matting baselines on challenging portrait cases.

ECCV 2026

ViewSplat: View-Adaptive Dynamic Gaussian Splatting for Feed-Forward Synthesis

Moonyeon Jeong, Seunggi Min, Suhyeon Lee, Hongje Seong

We present ViewSplat, a view-adaptive 3D Gaussian splat-ting network for novel view synthesis from unposed images. While recentfeed-forward 3D Gaussian splatting has significantly accelerated 3D scenereconstruction by bypassing per-scene optimization, a fundamental fidelitygap remains. We attribute this gap to the limited capacity of single-stepfeed-forward networks to regress static Gaussian primitives that sat-isfy all viewpoints. To address this limitation, we shift the paradigmfrom static primitive regression to view-adaptive splatting. Instead of arigid Gaussian representation, our pipeline learns a view-adaptive latentrepresentation. Specifically, ViewSplat initially predicts base Gaussianprimitives alongside the weights of scene-conditioned View MLPs. Duringrendering, these MLPs take target-view coordinates as input and predictview-dependent residual updates for each Gaussian attribute (i.e., 3Dposition, scale, rotation, opacity, and color). This mechanism, whichwe term view-adaptive splatting, allows each primitive to rectify initialestimation errors, effectively capturing high-fidelity appearances. Exten-sive experiments demonstrate that ViewSplat achieves state-of-the-artfidelity while maintaining fast inference and real-time rendering; our largebackbone variant runs at 15 FPS during inference and 90 FPS duringrendering.

ECCV 2026

OPAL: Orthonormal Prototype Alignment Learning for Interpretable Image Classification

Ilán Carretero, Gustavo ANGULO, Rocío Amor, Valery Naranjo

Prototypical part-based models provide explainable predic-tions by comparing input regions to learned prototypes. However, currentapproaches are burdened by complex, multi-stage training pipelines andheavily rely on auxiliary regularization to prevent prototype collapse. Toovercome these limitations, we introduce Orthonormal Prototype Align-ment Learning (OPAL), a single-stage, end-to-end framework that sim-plifies interpretable classification. Our approach anchors the latent spaceusing predefined orthonormal bases, embedding each class within a dedi-cated subspace spanned by fixed part-prototypes. To achieve precise partlocalization, OPAL enforces spatial competition across feature maps.This mechanism isolates sparse, discriminative regions, directing eachprototype to consistently attend to the same semantic concept across dif-ferent images. By framing classification as a direct representation align-ment task, our method eliminates the need for auxiliary losses. Extensiveexperiments on fine-grained benchmarks demonstrate that OPAL outper-forms both its non-interpretable counterparts and state-of-the-art part-prototype methods, delivering granular visual explanations by explicitlyrevealing the specific image regions driving every prediction. Code isavailable at https://github.com/ilancarretero/OPAL.

Keep browsing

More papers from top AI conferences

30+ top conferences, by conference, year, and topic.