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

FDM-MFVT: Few-step Sampling Diffusion Model for Mask-Free Virtual Try-On

Jiaxin Liu, XIAOYE LIANG, Lai Jiang, Jun Liu, Mai Xu

Image-based Virtual Try-On (IVTON) has greatly advancedthrough diffusion models, yet existing methods require many samplingsteps and depend on masks with costly auxiliary networks. In addition,the absence of large-scale mask-free paired datasets further limits thedevelopment of mask-free IVTON. We propose FDM-MFVT, a few-stepdiffusion model for mask-free IVTON, integrating an Outfit-aware NoiseOptimization Module (OANO) and an Instruction-driven Try-on Module(IDT) to enhance efficiency and flexibility.The OANO module initializesthe alignment space with noise using the input image and only needs 6steps to generate a higher-fidelity try-on image compared to 30 steps.TheIDT module uses virtual try-on prompts and efficient adaptation to gen-erate high-quality results from garment and person images alone. Wefurther introduce MFVT, a 30,000-pair mask-free IVTON dataset. Ex-periments show that FDM-MFVT achieves superior quantitative andqualitative results with fewer inference steps than mask-based and mask-free baseline methods.

ECCV 2026

ESC: Emotional Self-Correction for Reliable Vision-Language Models

Tien-Huy Nguyen, Nhat Nguyen, Nhat-Huy Nguyen, Hung Nguyen, Huy Nguyen, Thanh Huy Nguyen, Cuong Nguyen, Hoang Le, Dat Nguyen, Phat Huynh, Min Xu, Ulas Bagci

Vision-language models (VLMs) have achieved strong per-formance across diverse multimodal tasks, yet they remain vulnerable tounreliable reasoning. Existing self-correction methods mitigate these is-sues but typically rely on post-training or carefully engineered feedback,incurring high computational cost. In this work, we revisit this challengethrough the lens of emotional cues, asking whether they can activate la-tent self-correction behaviors in VLMs without additional training. Wefind that emotional signals serve as an effective trigger for self-correction, encouraging more cautious and reflective reason-ing. Motivated by this finding, we propose ESC (Emotional Self-Correction), a training-free self-correction framework. ESC introducesan external verifier that detects potentially incorrect initial responsesand injects emotional feedback to encourage model to reflect, and pro-duce a better revised response without additional training. Extensiveexperiments across safety, hallucination, vision-centric perception, andmultimodal reasoning benchmarks show that ESC consistently improvesreliability while preserving overall model utility. These results suggestthat emotion can function not only as an ability to be recognized, butalso as a practical control signal for scalable self-correction in VLMs. Wetherefore believe that ESC provides a strong foundation for anew reliable human-like, emotion-integrated research direction.Our project is publicly available at https://genai4e.github.io/ESC/.

ECCV 2026

Do Not Leave a Gap: Hallucination-Free Object Concealment in Vision-Language Models

Amira Guesmi, Muhammad Shafique

Vision-language models (VLMs) have recently shown remark-able capabilities in visual understanding and generation, but remainvulnerable to adversarial manipulations of visual content. Prior object-hiding attacks primarily rely on suppressing or blocking region-specificrepresentations, often creating semantic gaps that inadvertently inducehallucination, where models invent plausible but incorrect objects. In thiswork, we demonstrate that hallucination arises not from object absenceper se, but from semantic discontinuity introduced by such suppression-based attacks. We propose a new class of background-consistent objectconcealment attacks, which hide target objects by re-encoding their vi-sual representations to be statistically and semantically consistent withsurrounding background regions. Crucially, our approach preserves tokenstructure and attention flow, avoiding representational voids that trig-ger hallucination. We present a pixel-level optimization framework thatenforces background-consistent re-encoding across multiple transformerlayers while preserving global scene semantics. Extensive experimentson state-of-the-art vision-language models show that our method effec-tively conceals target objects while preserving up to 86% of non-targetobjects and reducing grounded hallucination by up to 3× compared toattention-suppression-based attacks. Qualitative results further confirmthat our approach maintains scene coherence and avoids spurious ob-ject insertion. Our findings highlight semantic continuity as a key factorin hallucination behavior and introduce a new direction for adversarialanalysis of generative multimodal models.

ECCV 2026

Do Multimodal LLMs Understand Intraoral Dental Data? Dataset, Platform, and Baselines

Luca Lumetti, Federico Rizzo, Francesca Cremonini, Ettore Candeloro, Lombardo Luca, Costantino Grana, Federico Bolelli

Progress in dental computer vision is limited by the absence of large-scale multimodal datasets that jointly capture 3D intraoral geometry and 2Dappearance across diverse clinical settings. Existing resources are typically uni-modal, which hinders robust cross-modal learning and generalization. We as-semble and release a multi-center dataset of 1,000 patients comprising 2,000registered upper/lower intraoral scans, 5,000 paired intraoral photographs, and2,403 clinician-authored reports. This combination links detailed 3D dental ge-ometry with complementary 2D evidence, supporting occlusal and orthodonticanalysis. Moreover, to enable scalable and privacy-preserving acquisition andannotation across distributed centers, we introduce an open platform that sup-ports multimodal ingestion and structured labeling. Experiments indicate thatstate-of-the-art multimodal models fail to generate clinically faithful reports, moti-vating geometry-aware adaptation. We therefore propose IOS-Qwen, which fuses aPointTransformer 3D encoder with Qwen3-VL to generate structured, point-cloud-conditioned reports. Together, the dataset, the platform, and the baselines establisha foundation for multimodal dental AI research. Code is publicly released.3

ECCV 2026

Diffusion Image Generation with Explicitly Modeling of Data Manifold Geometry

Duoduo Xue, Zhiyu Zhu, Junhui Hou

Image generative models aim to sample data points fromthe underlying data manifold, a task that requires learning and decod-ing a dense, low-dimensional, and compact parameterization space. Toachieve this, we propose the Data Manifold-aware Image diffusioN moDel(MIND), a novel framework that explicitly models manifold geometry byintegrating discrete patch tokenization into the score function of a con-tinuous diffusion model. This approach successfully leverages both thestructural quantification capabilities of discrete tokens and the parallelgeneration flexibility of continuous diffusion. Moreover, we enable end-to-end differentiable training via a novel soft top-k aggregation mechanismand introduce dual-branch high-frequency feature embedding layers toalleviate the spectral bias of transformer backbones on low-dimensionalinputs. Furthermore, for inference, we design a multi-stage transitionsampling scheme that dynamically adjusts the sampling scheme based ontimestep. Extensive experiments on ImageNet 256×256 demonstrate theeffectiveness of MIND. After 80-epoch training, our base model achievesan FID of 22.73 without guidance, nearly halving the 43.47 FID of thevanilla DiT-B/2 baseline. The proposed method reduces FID by 15.95and 9.06 on average compared with the baselines DiT and SiT, respec-tively. For image generation on ImageNet 256×256 with guidance, theproposed MIND-B with only 130M parameters achieves an FID of 2.06,surpassing LlamaGen-3B with 3.1B parameters. Our MIND-XL with715M parameters further reduces the FID to 1.95. Our MIND intro-duces a fresh perspective on diffusion-based image generation, pavingthe way for future research and innovation in this community. The codeis available here: https://github.com/xddgit/MIND.

ECCV 2026

CUST : Clustered Unit-level Similarity Transformer for Lightweight Image Super-Resolution

Jeongsoo Kim

Recently, Vision Transformer (ViT)-based models have exhibited remarkable performance in image super-resolution. However, the quadratic computational complexity of ViTs with respect to spatial resolution severely constrains their efficiency, leading to high latency and massive memory consumption. To alleviate this, various window-based attention mechanisms have been proposed; yet, they inherently compromise the long-range dependency modeling that is the primary advantage of ViTs. To overcome these limitations, we propose the Clustered Unitlevel Similarity Transformer (CUST), a novel architecture that efficiently integrates global and local information. Specifically, CUST enables each patch to aggregate and attend to similar patches within a broadened regional scope outside its local window, thereby capturing extensive contextual understanding. Furthermore, it employs overlapping attention windows to capture local dependencies, while explicitly extracting highfrequency details by computing the residual difference between the original features and their downsampled-upsampled counterparts. Comprehensive experiments demonstrate that our proposed model achieves a practical balance between computational efficiency and restoration performance. It achieves a lower memory footprint and faster inference speed compared to recent global context or lightweight models under realistic constraints. Code is available at https://github.com/jwgdmkj/CUST.

ECCV 2026

Concept-as-Tree: A Controllable Synthetic Data Framework Makes Stronger Personalized VLMs

Ruichuan An, Kai Zeng, Ming Lu, Sihan Yang, Renrui Zhang, Huitong Ji, Hao Liang, Wentao Zhang

Vision-Language Models (VLMs) have demonstrated excep-tional performance in various multi-modal tasks. Recently, there hasbeen an increasing interest in improving the personalization capabilitiesof VLMs. To better integrate user-provided concepts into VLMs, manymethods use positive and negative samples to fine-tune these models.However, the scarcity of user-provided positive samples and the low qual-ity of retrieved negative samples pose challenges for existing techniques.To reveal the relationship between sample and model performance, wesystematically investigate the amount and diversity impact of positiveand negative samples (easy and hard) on VLM personalization tasks.Based on the detailed analysis, we introduce Concept-as-Tree (CaT),which represents a concept as a tree structure, thereby enabling the datageneration of positive and negative samples with varying difficulty anddiversity, and can be easily extended to multi-concept scenarios. Witha well-designed data filtering strategy, our CaT framework can ensurethe quality of generated data, constituting a powerful pipeline. We per-form thorough experiments with various VLM personalization baselinesto assess the effectiveness of the pipeline, alleviating the lack of positivesamples and the low quality of negative samples. Our results demon-strate that CaT equipped with the proposed data filter significantlyenhances the capabilities of VLMs across personalization benchmarks.To the best of our knowledge, this work is the first controllable syn-thetic data pipeline for VLM personalization. The code is released athttps://github.com/zengkaiya/CaT.

ECCV 2026

COLA: Continual Orthogonal Low-Rank Adaptation for Class-Incremental Learning

Monu Nagar, Debasis Das

Recent advances in Continual Learning (CL) have adoptedfoundation models with Low-Rank Adaptation (LoRA) or prompt tuning,utilizing pre-trained representations to achieve more efficient and adapt-able learning. However, these methods suffer from two key challenges.First, they learn task-specific adapters, resulting in unbounded parameterexpansion as task sequences grow. Second, they maintain replay buffersor store feature representations from previous tasks, which introducessubstantial memory overhead and potential privacy risks. To address theselimitations, we propose Continual Orthogonal Low-Rank Adaptation(COLA), a novel rehearsal-free, parameter-efficient framework for CL.COLA integrates LoRA’s low-rank adaptation with an Oja-inspired learn-ing rule that incrementally approximates the dominant eigenstructure ofthe feature covariance across tasks. This mechanism continuously tracksthe principal directions of prior task knowledge and progressively reducesrepresentational interference between new and previously learned tasksubspaces. By dynamically adapting a shared low-rank subspace usingcovariance updates, COLA achieves stable feature projection and con-tinual knowledge integration without additional parameter expansionor memory replay. Extensive experiments on class-incremental bench-marks such as ImageNet-R, ImageNet-A, and CUB200 demonstrate thatCOLA effectively mitigates catastrophic forgetting while maintainingminimal memory overhead and strong generalization, outperforming ex-isting LoRA-based and prompt-based methods. The code is available at:https://github.com/autovisionproject/COLA.

ECCV 2026

Stable and Scalable Bundle Adjustment of Holistic 3D Structures

Shaohui Liu, Remi Pautrat, Daniel Barath, Richard Hartley, Viktor Larsson, Marc Pollefeys

Bundle Adjustment (BA) is a cornerstone of 3D computervision and has benefited from decades of advances in sparse optimiza-tion and numerical methods. It was originally developed for jointly op-timizing camera intrinsics, poses and sparse 3D points. While exten-sions incorporate lines and other primitives, integrating richer geometricstructures such as parallelism, coplanarity, or wireframes often intro-duces significantly increased computational cost and reduced numericalstability. In this paper, we propose a unified framework that extendsbundle adjustment to jointly optimize geometric features and higher-order relations. We first introduce a taxonomy that distinguishes scal-able geometric features with direct 2D measurements (e.g., points andlines), from groups encoding higher-order relations (e.g., coplanarity, par-allelism, etc.), where we show that groups can be modeled as camera-likeentities within the bundle adjustment framework. Building on this formu-lation, we propose that both group constraints and cross-feature relations(i.e., point–line associations) can be expressed through 2D reprojectionmeasurements. By formulating group-induced and cross-feature repro-jection errors, we preserve the sparsity structure of classical point-basedBA under Schur elimination, while avoiding direct 3D regularization thatdegrades the conditioning and stability. Experiments on both real-worldand synthetic datasets demonstrate runtime performance comparableto classical point-only bundle adjustment, while producing significantlyricher 3D structures and improved geometric accuracy.

ECCV 2026

Beyond Common Sense: Grounding Logical Anomaly Detection in Inspection Criteria

Yuanze Li, Shihao Yuan, Zimeng Zhu, Ming Liu, Wangmeng Zuo, Guangming Shi

Multimodal large language models have significantly advancedzero-shot industrial anomaly detection, yet they remain highly constrainedwhen identifying logical anomalies. Fundamentally, logical anomalies arisefrom discrepancies between inspected products and their predefined con-straints. Current detection paradigms typically operate without accessto these explicit standards, forcing models to rely on pretrained com-mon sense rather than strict logical rules. To address this limitation,we introduce SCAN, a Systematic Criteria-driven ANomaly inspec-tion framework optimized through a two-stage post-training pipeline.We first employ supervised finetuning based on rule-by-rule Chain-of-Thought (CoT) reasoning to systematically verify explicit inspection cri-teria. We then apply reinforcement learning equipped with a rule-levelcredit assignment mechanism to enforce precise anomaly attribution andovercome severe errors caused by sparse feedback. However, such criteria-driven training demands large-scale logical anomaly data paired with ex-plicit inspection rules, yet existing datasets remain largely skewed towardstructural defects. To bridge this gap, we propose FLAW, a Fine-grainedLogical Anomaly dataset With criteria. This dataset leverages imagesynthesis and logical data retrieval to endow images with rich object-levelmetadata, enabling the construction of diverse inspection standards andlogical violations. Extensive experiments on MVTec LOCO demonstratethat our 8B-parameter SCAN model improves upon its base model by7.8% and establishes a new state of the art, outperforming GPT5.2-Chatand Gemini-3.1-Pro by 4.8% and 1.0%, respectively. Source code anddata will be publicly available at https://github.com/tzjtatata/SCAN.

ECCV 2026

AV2T-Gen: Aerial Visible to Thermal Generation with Environment and Vehicle State Guidance

Kun Yang, Yuxiang Liu, Yihan Wang, Shen Yan, Maojun Zhang, Yu Liu, Xue Wang, Qing Wang

Accurate visible-to-thermal image generation holds signifi-cant application value in fields such as vehicle detection and autonomousdriving. However, existing methods often suffer from the problem thatthe generated results are inconsistent with environmental conditions andvehicle states (e.g., moving/static). Moreover, the scarcity of synchro-nized and calibrated visible-thermal image pairs has further hinderedthe development of these fields. We propose AV2T-Gen, an innovativeenvironment and vehicle guided visible-to-thermal generation frameworkbuilt upon the Instruct-Pix2Pix Diffusion Models for aerial imagery,which distinguishes itself from existing methods by explicitly incorpo-rating environment parameters and vehicle state information for thefirst time. The framework first maps images to the latent space usinga VAE to process image features efficiently, and then generates direc-tional features for vehicles in different environmental conditions usingmultimodal feature information. Meanwhile, we also proposed a visible-thermal paired dataset with environmental conditions and vehicle states.Experimental results have demonstrated that the generated thermal im-ages outperform existing approaches in terms of physical realism andvehicle detail discrimination significantly, with more accurate reflectionof environmental and vehicle characteristics. The code is available athttps://github.com/NPU-CVPG/AV2T.

ECCV 2026

White Aggregation and Restoration for Few-shot 3D Point Cloud Semantic Segmentation

Jiyun Im, SuBeen Lee, Miso Lee, Jae-Pil Heo

Few-shot 3D Point Cloud Semantic Segmentation (FS-PCS)aims to predict per-point labels for an unlabeled point cloud, given onlya few labeled examples. To extract representations from the limited la-beled set, existing methods have constructed prototypes with FarthestPoint Sampling (FPS). However, we found that this convention resultsin performance instability due to its sensitivity to FPS-induced varia-tions, while the prototype generation process remains underexplored inthe field. This motivates us to investigate deterministic prototype gen-eration method based on attention mechanism. Despite its potential, wefound that vanilla attention module suffers from the distributional gapbetween prototypical tokens and support features. To overcome this, weprovide a simple approach, White Aggregation and Restoration Module(WARM), which resolves the misalignment by wrapping cross-attentionwith whitening and coloring transformations. Specifically, whitening alig-ns the features to tokens before the attention process, and coloring subse-quently restores the original distribution to the attended tokens. This de-sign enables robust attention, thereby generating prototypes that capturethe semantic relationships in support features. WARM achieves state-of-the-art performance with a significant margin on the S3DIS dataset, andcompetitive performance on the ScanNet dataset. Further experimentsdemonstrate its effectiveness in deterministic prototype generation. Codeis publicly available at: https://github.com/JiyunIm00/WARM.git

ECCV 2026

What Images Cannot Say: Language-Guided Olfactory Representation Learning

Eleftherios Tsonis, Xi Wang, Vicky Kalogeiton

Images tell us what a scene looks like, but rarely what itwould feel like to be there. While recent datasets pair visual sceneswith electronic-nose measurements, aligning smell signals with imagesremains challenging because many olfactory cues arise from contextualenvironmental factors that are not directly visible in pixels. We intro-duce SCENT, a multimodal framework that uses language guidance asa semantic bridge between vision and olfaction. Our approach leveragesVision-Language Models (VLMs) to generate scene descriptors captur-ing objects, environmental context, and plausible ambient smell cuessuggested by the visual scene. These descriptors provide semantic guid-ance for learning olfactory representations. We train a smell encoderthat maps electronic-nose signals into a shared embedding space alignedwith both visual and textual representations, and introduce a language-guided latent decomposition that separates object-specific odors fromcontextual environmental contributions. Experiments on the New YorkSmells dataset demonstrate that SCENT significantly improves cross-modal retrieval compared to vision-only baselines, achieving state-of-the-art performance on smell-to-image and smell-to-text retrieval tasks. Inaddition, our framework produces interpretable olfactory representationsthat enable the disentanglement of complex smell mixtures. Our resultsreveal the importance of contextual semantic information for groundingolfactory perception in multimodal learning and pave the way for futureresearch in this area.

ECCV 2026

VLMSysTrojan: Stealthy System-Aware Backdoor Attacks Against Vision-Language Models

Yezheng Cheng, Zexin Li, Jiaqi Wu, Zhihong Zhang, Simin Chen

Vision–Language Models (VLMs) are increasingly deployedin real-world applications. To ensure efficient inference, these deploy-ments typically rely on specialized system kernels (e.g., CUDA or cuDNN)for acceleration. However, existing system-level research on VLMs has fo-cused primarily on improving kernel efficiency and performance, whilelargely overlooking their potential impact on model security and ro-bustness. In particular, the security implications of inconsistencies be-tween training and inference kernels remain underexplored. To addressthis gap, we present the first systematic study revealing a new classof vulnerabilities: system-level backdoor attacks that exploit floating-point inconsistencies across training and inference kernels. We introduceVLMSysTrojan, a red-teaming framework that constructs models behav-ing benignly under standard training kernels but exhibiting backdooredbehavior when executed on specific target inference kernels. Specifically,VLMSysTrojan produces models whose backdoor triggers remain inactiveunder standard kernels and successfully evade three state-of-the-art back-door detection methods. Yet, when deployed on a target kernel, kernel-specific floating-point behavior activates the backdoor. Empirically, thecompromised models achieve a 99% attack success rate on triggered in-puts while maintaining normal accuracy on clean inputs. Moreover, theattack generalizes across multiple kernel configurations, exposing a previ-ously unrecognized risk in VLM deployment and highlighting the urgentneed for kernel-aware robustness analysis techniques.

ECCV 2026

VIPS: Vehicle-Infrastructure Cooperative Planning Benchmark via Pseudo-Simulation

Hoonhee Cho, Jae-Young Kang, Giwon Lee, Hyemin Yang, Heejun Park, Kuk-Jin Yoon

End-to-end autonomous driving in urban environments re-quires robust decision-making under partial observability and complexmulti-agent interactions. Severe occlusions and dense traffic at inter-sections limit the perception capability of single-agent systems, moti-vating recent efforts on Vehicle-to-Infrastructure (V2I) cooperation forperception and planning. However, existing evaluation protocols face afundamental trade-off: open-loop evaluation fails to capture error accu-mulation and recovery from deviations, while closed-loop evaluation iscostly, difficult to scale, and often relies on simulated environments thatmay suffer from domain gaps. To bridge this gap, we propose VIPS, abenchmark for cooperative autonomous driving in V2I settings based onpseudo-simulation. VIPS extends pseudo-simulation by integrating ve-hicle and infrastructure observations. This enables scalable yet realisticevaluation of robustness and error propagation without full simulation.We further present CoS-V2X, a cooperative planning framework basedon sparse representations. CoS-V2X models vehicle–infrastructure inter-actions using compact features for efficient communication and robustdecision-making under heterogeneous observations. Code and dataset areavailable at https://vips2026.github.io.

ECCV 2026

Towards Spatial Supersensing in the Wild

Tianjun Gu, Tianyu Xin, Kuan Zhang, Bowen Yang, Yinan Han, Peize Li, Yucheng Lu, Jianhang Liu, Xinran Zhang, Kok Chung Chua, Qiyue Zhao, Qinlei Xie, Yupeng Chen, Marco Pavone, Yiming Li

Humans can efficiently parse continuous sensory streams,from hours to years, scaffolding an internal world model that groundsspatial reasoning and prediction. To mimic this capacity, spatial super-sensing [59] challenges multimodal models to move beyond linguistic un-derstanding toward true world modeling. However, their benchmark re-lies on synthetic long videos—formed by concatenating random shortclips—and is mostly limited to household scenes, leaving real-world con-tinuity and diversity underexplored. To address the gap, we introduceVSI-Super-Wild, a large-scale spatial supersensing benchmark builtfrom genuinely long, in-the-wild videos across diverse scenarios. Notably,inspired by cognitive studies on how humans structure experience, wesystematically probe the full triad of world state: the agent (observer),objects (scene items), and the environment (places and global layout).* †Equal contribution. Corresponding author.In total, VSI-Super-Wild comprises 442 real-world long-form videosacross 8 scene categories and 6,980 human-verified question-answer pairs.Evaluating multimodal models on VSI-Super-Wild exposes a funda-mental disconnect: despite advances in static image understanding, mod-els consistently fail at tasks that require coherent world-state trackingover time. We characterize how performance degrades with world-statecomplexity and temporal horizon, and diagnose four failure modes: spa-tial collapse, semantic shortcuts, insufficient update, and instance con-fusion. This taxonomy reveals that models lack the mechanisms to bindobjects, agents, and environments into a unified spatial world model—afundamental gap that defines the path forward for spatial supersensing.

ECCV 2026

The Devil Is in the Dark Pixels: Toward Brightness Bias-Robust Denoising

Sungjun Cho, Zhuangzhuang Chen, Xiaomeng Li

In this paper, we reveal an important yet overlooked problem in image denoising: under signal-dependent camera noise models, dark regions suffer from inherently low Signal-to-Noise Ratio (SNR), as signal intensity decays far faster than noise variance diminishes, making detail recovery in dark areas fundamentally challenging. Yet rather than compensating for this difficulty, MSE-trained denoisers exacerbate it—reconstructing dark pixels up to 6× worse relative to their perband noise floor. This bias stems from two compounding factors: signaldependent noise inflates bright-pixel residuals, and the network’s Jacobian norm increases monotonically with brightness. Together, these cause bright regions to chronically dominate gradient updates at the expense of dark ones. To this end, we propose Brightness Bias-Robust Denoising (BBRD), a drop-in replacement for MSE loss that partitions pixels into brightness bands, normalizes per-band error by empirical noise variance, and applies Group Distributionally Robust Optimization (Group-DRO) to dynamically upweight whichever band is currently worst, with zero additional parameters or inference cost. Across 8 architectures and 2 datasets in our experiments, BBRD is the only method among 13 tested alternatives that improves each brightness band simultaneously, achieving up to +0.45 dB on dark bands, +0.32 dB on bright bands, and +0.65 dB aggregate Peak Signal-to-Noise Ratio (PSNR) on SIDD, with the largest per-band gains in the darkest regions where detail recovery matters most. Code is available at https://github.com/xmedlab/BBRD.

ECCV 2026

Tesselating The Earth

Daniel Cher, Hamza Iqbal, Eric Xing, Brian Wei, Nathan Jacobs

Geolocation encoders, which map geographic coordinates tolearned representations, are emerging as an effective means of capturingvisual and non-visual characteristics from a latitude-longitude pair alone.However, existing approaches project coordinates onto fixed bases (e.g.,spherical harmonics), allocating representational capacity uniformly anddevoting equal resources to the open ocean and to a developing city. Weintroduce Tessellating the Earth (TTE), a location encoder built fromlearnable Spherical Voronoi partitions that concentrates representationalcapacity where it is needed in a fully differentiable, end-to-end manner.Each Voronoi site carries its own embedding and migrates during train-ing toward discriminative areas. To bridge the gap between local spatialstructure and global semantic understanding, we introduce global seman-tic tokens: a set of shared learnable concept tokens that distill semanticknowledge from the satellite imagery into a compact vocabulary the lo-cation encoder can reference at inference, enabling geographically distantsites covering similar environments to share semantics. TTE sets a newstate of the art for location encoders across a suite of geospatial classifi-cation and regression tasks, and achieves the strongest results when usedas a geographic prior for fine-grained species classification on iNaturalist-2018. Code, and weights are available at https://github.com/mvrl/TTE.

ECCV 2026

Task Alignment: A simple and effective proxy for model merging in computer vision

Pau de Jorge Aranda*, Cesar De Souza, Björn Michele*, Mert Bulent Sariyildiz, Philippe Weinzaepfel, Florent Perronnin, Diane Larlus, Yannis Kalantidis

Efficiently merging several models fine-tuned for differenttasks, but stemming from the same pretrained base model, is of greatpractical interest. Despite extensive prior work, most evaluations of modelmerging in computer vision are restricted to image classification usingCLIP, where different classification datasets define different tasks. In thiswork, our goal is to make model merging more practical and show itsrelevance on challenging scenarios beyond this specific setting. In mostvision scenarios, different tasks rely on trainable and usually heteroge-neous decoders. Differently from previous studies with frozen decoders,where merged models can be evaluated right away, the non-trivial cost ofdecoder training renders hyperparameter selection based on downstreamperformance impractical. To address this, we introduce the task align-ment proxy, and show how it can be used to speed up hyperparameterselection by orders of magnitude while retaining performance. Equippedwith the task alignment proxy, we extend the applicability of modelmerging to multi-task vision models beyond CLIP-based classification.

ECCV 2026

SynLF: Zero-Shot Metric Depth from Light Field Cameras via Physics-Grounded Synthesis

Zhexuan Cao, Yuduo Guo, Peisheng Ding, Zhan Shi, Hui Qiao

Accurate monocular metric depth estimation remains inher-ently ill-posed due to scale ambiguity. Light field cameras mitigate thisambiguity by capturing micro-baseline disparities within a single lensto recover metric depth, but current learning-based solutions are lim-ited by scarce training data and often overfit Lambertian appearancepatterns, restricting real-world robustness. To address this, we proposeSynLF, which alleviates data scarcity via physics-grounded light field(PG-LF) synthesis and performs robust real-world metric depth esti-mation through VisDepth, a prior-initialized iterative estimator with ex-plicit multi-view occlusion reasoning. During training, we synthesize lightfield supervision on-the-fly from RGB-D via physics-grounded light fieldsynthesis, including geometry-consistent view synthesis, depth-dependentdefocus, and non-Lambertian perturbations. At inference, the model isdirectly evaluated on real light field captures without fine-tuning. Ex-tensive experiments demonstrate that SynLF generalizes zero-shot toour custom real-world light field dataset, which features highly chal-lenging conditions such as transparency, reflections, specular highlights,and textureless regions. It achieves 30 mm absolute and 1.4% relativeerrors across 0.5–2.5 m, demonstrating that physics-grounded LF syn-thesis from large-scale RGB-D provides a scalable and storage-efficientalternative to conventional light field acquisition for accurate single-lensmetric depth estimation.

Keep browsing

More papers from top AI conferences

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