3D
-
SemEllipsoid: Semantic-aware Shape Abstraction with Ellipsoid
2026-10-30 17:59:14Kuankuan Cheng, Haocheng Yuan, Changjian LiAbstract
Representing 3D shapes using a set of primitives - known as shape abstraction - is an actively studied problem in computer graphics with wide applications in shape understanding, analysis, and simulation due to the compactness and computational efficiency of the primitive-based representation. Prior work has primarily focused on geometry-aware shape abstraction, where the objective is to approximate the input shape with a minimal set of primitives while preserving geometric fidelity. However, these methods often overlook semantic consistency, inevitably leading to primitives that span across multiple semantic parts. Such inconsistency can cause undesirable artifacts in downstream tasks such as part-level shape editing or manipulation. In this paper, we introduce a new task - semantic-aware shape abstraction, where the goal is to produce abstractions that are both geometrically faithful and semantically consistent, ensuring that each primitive adheres strictly to a single semantic part without crossing semantic boundaries. We propose SemEllipsoid, a novel method to address this problem using ellipsoidal primitives. The core idea is to inject semantic constraints into the primitive optimization process leveraging the availability of semantic cues from large-scale foundation models. We design a fully differentiable projection-blending-evaluation scheme to measure semantic divergence against input semantic masks, and formulate an alternating optimization strategy to iteratively solve for optimal semantic assignment and primitive configurations. Our method is non-learning-based, requiring no data collection or network training, and generalizes well to shapes from arbitrary categories. It also supports different levels of abstraction according to user demand. Extensive experiments and evaluations validate the effectiveness of our approach across several 3D shape collections.
Abstract (translated)
URL
https://arxiv.org/abs/2610.42145
PDF
-
ReLift: Dense 3D Tracking as Lifted Trajectory Repair
2026-10-30 17:58:33Assaf Singer, Maroun Khriesh, Noam Rotstein, Yoni Kasten, Or LitanyAbstract
Dense 3D tracking from monocular video requires establishing point correspondences over time, estimating their depth, and inferring their trajectories during occlusion. Existing methods typically learn these tasks jointly, a demanding objective given the scarcity of 4D supervision. We show that strong pretrained 2D trackers and 3D geometry models provide an effective foundation: lifted trajectories, with occluded segments simply interpolated, already rival dedicated 3D trackers using the same geometry. However, lifted trajectories inherit correspondence and depth errors, and under occlusion, their depth comes from the occluder rather than the tracked point. We introduce ReLift, an efficient feed-forward framework that recasts dense 3D tracking as geometric trajectory repair and completion. At its core, a trajectory corrector alternates attention along and across trajectories to preserve reliable observations, repair corrupted positions, and complete missing motion. Using only geometric inputs allows us to train directly on 3D trajectories with simulated lifting errors, decoupling training from images and from any specific tracker or geometry model. Concentrating 4D supervision on trajectory refinement yields a reusable model that consistently improves lifted trajectories across trackers and geometry providers without retraining, so it benefits directly from future advances in either component. On WorldTrack, ReLift achieves state-of-the-art mean accuracy, improving on the strongest baselines by 2.1 APD on all points and 3.0 on dynamic points, with faster configurations remaining competitive with leading 3D trackers while running faster than 4RC. Visit our supplementry page for more examples and code.
Abstract (translated)
URL
https://arxiv.org/abs/2610.42136
PDF
-
Rethinking Asymmetry in Self-Supervised 3D Geometry Learning
2026-10-30 17:38:39Gyeongjin Kang, Freeman Cheng, Botao Ye, Weijie Lyu, Ming-Hsuan Yang, Eunbyung ParkAbstract
Self-supervised learning relies on asymmetry between what a model observes and what it must explain. Recent methods learn 3D geometry from unposed images by predicting cameras and scene structure, then rendering held-out views. To avoid trivial solutions, they impose this asymmetry in the input space, restricting how views interact through separate branches or masked attention. We revisit this design with geometric accuracy, rather than rendering quality, as the main objective. When a held-out view lies between context views, the context already constrains the geometry it observes. We therefore impose the asymmetry in the 3D representation space, excluding held-out views only from the reconstructed scene while a shared backbone uses all views for correspondence. This improves depth and pose estimation over input-space designs. We further exploit view density: sparse and dense observations depict the same scene but provide different amounts of correspondence. Aligning sparse-context features with dense-context features supplies a geometric training signal alongside rendering, without a pretrained geometric teacher. We additionally align features with a frozen vision foundation model, which brings further gains. None of these signals requires geometric annotation, and the resulting model improves geometry estimation over prior self-supervised methods. Controlled experiments isolate the contribution of each component, and self-supervised pretraining also improves subsequent supervised training.
Abstract (translated)
URL
https://arxiv.org/abs/2610.42092
PDF
-
Denoising Multirobot Trajectories
2026-10-30 15:13:03Yuhao Zhang, Keisuke Okumura, Ajay Shankar, Amanda ProrokAbstract
Multi-robot trajectory planning is a fundamental problem in multi-robot coordination but remains computationally challenging due to its nonconvex, multimodal, and high-dimensional nature. This work builds upon D4orm, a dynamics-aware diffusion-denoising framework, and develops a family of planning architectures for diverse operational requirements. Unlike conventional numerical optimization methods, D4orm employs sampling-based optimization to generate solution trajectories through massively parallel sampling, leveraging modern computing architectures such as GPUs. Its diffusion-denoising structure iteratively optimizes deformations to candidate control trajectories, providing an efficient and versatile paradigm for generating kinodynamically feasible and conflict-free trajectories. Using D4orm as the building block for advanced planners, we present a decoupled planner for improved scalability, an online receding-horizon planner with feedback control, and a distributed planner for resource-constrained settings. Evaluations with differential-drive and holonomic robots in 2D and 3D environments demonstrate that D4orm-based approaches find high-quality solutions faster and more reliably than other sampling-based optimization methods, such as MPPI, as well as a learned diffusion-model-based method. We further demonstrate zero-shot deployment on ten real quadrotors with obstacles, large-scale deconfliction with 100 simulated robots, and fully onboard distributed `lifelong' operation with six ground robots. Overall, these results establish diffusion denoising as a scalable and reliable framework for multi-robot coordination. Code and video: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2609.35651
PDF
-
ObjectGuidedMap: An Object-Aware Model for Robust Bird's Eye View Map Prediction in High-Occlusion Scenarios
2026-10-30 14:51:06Christian Bohn, Harsh Yadav, Klaus Friedrichs, Hasan Tercan, Tobias MeisenAbstract
Accurate BEV map prediction is a critical component of autonomous driving systems, enabling downstream planning and control. Recent approaches such as Mask2Map achieve strong performance by predicting vectorized, instance-level map elements directly from multi-view images. However, existing BEV map prediction models operate largely independently of object-level scene understanding, despite the fact that traffic participants frequently occlude or implicitly encode information about road geometry. In this work, we propose a multitask architecture that jointly predicts BEV maps and 3D objects, explicitly enabling information exchange between the two tasks. Building upon Mask2Map, we introduce an object detection branch and incorporate cross-task feature fusion at both the BEV feature level and the instance-query level via deformable attention and cross-attention mechanisms. This design allows map instances to leverage object context while remaining robust to occlusions caused by traffic participants. Experimental results on the nuScenes and Argoverse 2 datasets demonstrate that our approach consistently improves BEV map prediction accuracy, with particularly significant gains of 21% mAP in scenes containing heavy object occlusion.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41834
PDF
-
WorkIng: A Real-World and Synthetic Highway Work-Zone Benchmark Dataset for Autonomous Navigation
2026-10-30 14:11:34Xujun Xie, Dominik R\"o{\ss}le, Adithya Mohan, Daniel Cremers, Torsten Sch\"onAbstract
Highway work zones are critical edge cases for autonomous vehicles (AVs), combining atypical geometries and jurisdiction-specific regulations that existing benchmarks fail to capture: real-world datasets lack 360° multi-modal, full-scene 3D work zone annotation, and no single recorded site can span the layouts and regulatory variants needed for generalizable models. We present WorkIng, a dual-domain benchmark and training resource. WorkIng-Highway provides a 360° multi-modal work zone benchmark with synchronized 128-channel LiDAR, six RGB cameras, and RTK-GPS across 32 km, yielding 12,000+ annotated frames and 227,399 tracked 3D bounding boxes. WorkIng-CARLA complements this with procedural simulation spanning 24 regulation-aware scenarios across six canonical topologies and six international traffic standards, populated by 100+ custom 3D assets. Benchmarking representative 3D detection and end-to-end planning architectures shows that real-world data remains the primary performance driver, and that scenario-stratified synthetic augmentation improves architectures across 2D recognition, 3D detection, and trajectory planning. Resources: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.41772
PDF
-
Enhancing Medical Lesion Segmentation via Several Simple Modules
2026-10-30 12:26:57Ruichen Li, Zhao YaoAbstract
This work introduces three lightweight enhancement modules for medical lesion segmentation to improve clinical applicability, particularly under data-scarce and computationally constrained environments. Current mainstream methods struggle with the lack of high-quality annotations for rare pathologies, which often leads to severe morphological distortions during inference, despite achieving seemingly ideal DSC scores. Furthermore, the isolated development of individual segmentation tasks restricts cross-domain methodological innovation, underscoring the need for leveraging shared knowledge across different lesions within the same anatomical region. However, constrained access to historical data and annotations hinders efficient knowledge extraction and triggers catastrophic forgetting when adapting to new classes. To address these challenges, we first propose DT-Morph Combined Loss with continuous spatial pooling to optimize differentiable topological structures and boundary continuity. Second, we introduce 3D Group Prompt, which extracts domain-universal knowledge via curated volumetric image-mask alignment prompts together with text prompts built on SAT baseline. Finally, we replace conventional output layers with Many-to-One Conditional Segmentation Head equipped with synonym-resolved routing, which brings negligible computational cost and enables fast incremental learning for new categories without sacrificing performance on old tasks. Ablation studies on abdominal and extra-abdominal datasets covering both organ and lesion segmentation verify the efficacy of our modules.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41652
PDF
-
Direct Action-Manifold Modeling for Efficient and Robust Bimanual Manipulation
2026-10-30 12:17:46Hongren Wang, Mingyu You, Zihan Li, Zhipeng Wang, Shuo Jiang, Bin HeAbstract
Diffusion and flow matching visuomotor policies require iterative denoising or numerical integration for multimodal action generation, introducing latency and execution mismatch. We instead consider a different formulation: conditioned on vision and proprioception, demonstrated bimanual action chunks concentrate in a structured observation-dependent region of joint-action space, making the action target more direct to model than the generative process. Based on this view, we propose the Direct Manifold Action Transformer (DMAT), which reformulates action generation as the direct modeling of clean, observation-conditioned joint-space action chunks from stochastic interpolants, thereby enabling single-step inference without relying on iterative generative inference or auxiliary trajectory-straightening procedures. Within DMAT, local 3D geometry anchors predictions to spatial structure, while State-Consistent Spectral-Time Refinement (STR) reconciles predicted chunks with proprioception and suppresses high-frequency residuals. Across 567 standard and out-of-distribution (OOD) trials per method, DMAT outperformed Improved 3D Diffusion Policy (IDP3) and DiTFlow3D in aggregate success rate, achieving relative improvements of 129.1% and 95.7%, respectively, while generating action chunks 2.06x and 1.34x faster and remaining robust under repeated 50 N interventions. These results pave the way for a novel generative paradigm for efficient and robust bimanual manipulation.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41643
PDF
-
Patch-Based Deep Kernel Learning for Data Assimilation and Bias Correction with Extremely Sparse Observations
2026-10-30 11:29:02Emanuele Mele, Angelo Coluccia, Massimo Cafaro, Italo Epicoco, Paola NassisiAbstract
Data Assimilation (DA) reconstructs the state of a spatio-temporal system by combining sensor observations with the output of a numerical model. Since model states are high-dimensional while observations cover only a small fraction of the domain, DA is a challenging large-scale estimation problem, central to weather forecasting and oceanography. Classical variational methods such as 3D-Var rely on a hand-crafted background error covariance that assumes unbiased errors, and therefore cannot capture state-dependent correlations or systematic model biases. Moreover, their cost grows with the size of the domain. Deep learning (DL) methods can perform DA and bias correction (BC) non-linearly, but typically offer limited interpretability and no calibrated uncertainty estimates. We propose a hybrid framework that couples a convolutional neural network (CNN) with a learnable-kernel Gaussian Process (GP) to jointly perform DA and BC from extremely sparse observations. The model is trained end-to-end by maximizing the marginal log-likelihood, learns a data-driven covariance over the whole grid, and provides both the Maximum A Posteriori analysis and its uncertainty by design. A patch-based strategy reduces the $O(N^3)$ training cost of exact GPs over $N$ grid points to $O(N)$ for a fixed patch size. Inference remains exact and only requires solving a linear system whose $O(m^3)$ cost depends on the number of observations $m \ll N$, not on the domain. The learned covariance can also replace the hand-crafted one in 3D-Var, improving its accuracy without changing the variational formulation. Experiments on sea surface temperature and salinity, with both synthetic and real, extremely sparse in-situ observations, show that the framework outperforms 3D-Var, a classical GP and a competing DL model, while reducing inference time by more than 98% with respect to 3D-Var.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41596
PDF
-
Self-Localization of Distributed Microphone Arrays Using One Source and Acoustic Reflections
2026-10-30 10:43:55Thomas Guffroy (MACARON, UMRAE), Jean-Daniel Pascal Prieto (UMRAE, MACARON), C\'edric Foy (UMRAE), Antoine Deleforge (MACARON)Abstract
We present a method to estimate the relative position and orientation of two non-coplanar microphone arrays using a room impulse response (RIR) from a single source. The approach exploits early acoustic reflections, retrieved by each array in the form of a 3D image-source point cloud, using a recently-proposed image-source localization technique. The task is then reduced to aligning the two unlabeled point clouds, which is made difficult by their inherent symmetries and by spurious, missing and noisy estimates. To this aim we introduce ARCTIC, a point-cloud registration algorithm that performs a ranked search over congruent triangles anchored on the true source. Simulated and real experiments show that ARCTIC strongly outperforms two baselines, achieving mean angular and radial self-localization errors below 5 and 10 cm across 36 real RIRs measured in 6 different rooms.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41547
PDF
-
VoxLution: Evolving Voxels for High-Fidelity 3D Scene Generation
2026-10-30 09:55:52Daoyi Gao, Xinyi Zhang, Alexey Bokhovkin, Lei Li, Angela DaiAbstract
High-resolution 3D indoor scene generation is crucial for immersive applications such as virtual reality, spatial computing, and robotics simulation. We introduce VoxLution, a coarse-to-fine transformer-based approach that generates high-fidelity indoor scenes with rich details, overcoming the overly smooth geometry typical of existing generative models. First, we generate a low-resolution spatial layout; then, based on this structural information, we progressively refine the scene geometry locally in chunks. Key to our approach is formulating 3D super-resolution as a sequence modeling problem over multi-resolution sparse voxel grids. To achieve this, we introduce a training-free, resolution-invariant tokenization scheme that compactly encodes the geometric information of eight octants into a single decimal number. We formulate the transformer sequence to include multiple resolutions, allowing the coarse layout to guide the generation of fine local details. To ensure coherence across local chunks, we incorporate adjacent chunk boundary information directly into the transformer sequencing. VoxLution achieves scene generation at a finer voxel resolution than existing baselines, demonstrating state-of-the-art performance on standard benchmarks with 59.6% lower FPD and 51.1% lower FID on 3D-FRONT, and 70.5% lower FPD and 73.8% lower FID on Scannet++.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41501
PDF
-
Task-Performance Routing for Multi-Teacher Distillation in CT Medical Image Segmentation
2026-10-30 07:27:29Jiaye Yang, Yao Liu, Die Dai, Hanwen Zhang, Peiyuan Jiang, Qiao Liu, Yutong Xie, Peng WangAbstract
Pre-trained models offer complementary strengths in CT organ segmentation, but their relative accuracy varies across organs and spatial regions. Multi-teacher knowledge distillation can combine these strengths in a single student, yet global or instance-level teacher weights overlook variation within a 3D volume. We propose Task-Performance Routing (TPR), which uses region-wise segmentation errors to determine teacher weights during training without learned routing parameters. In hard regions, TPR favors teachers with lower prediction errors; in easy regions, it uses uniform weights to combine their knowledge. A hierarchical spatial partition provides fine-grained routing in shallow encoder stages and coarse-grained routing in deep stages. Experiments on BTCV, MSD-Liver, and MSD-Pancreas with two teacher pairs and two backbone architectures show higher Dice scores than the individual teachers, adaptive ensembles, and evaluated multi-teacher distillation baselines. On MSD-Pancreas, TPR achieves Dice scores of 65.40% with SwinUNETR and 68.25% with U-Net, compared with 64.38% and 67.45% for the corresponding adaptive ensembles. Inference uses the student alone, with no routing or teacher overhead. Code is available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41362
PDF
-
Transferring Peptide-Target Interaction Patterns to Small Molecules by Joint Guidance for Diffusion and Flow Matching
2026-10-30 06:33:09Jian Lee, Changyun Cho, Sun KimAbstract
Generating small molecules to target protein-protein interactions (PPIs) remains one of the most challenging yet high-reward frontiers in drug discovery. Unlike traditional small-molecule drugs that bind to deep, well-defined pockets, PPI interfaces often comprise broad, shallow binding surfaces. A natural strategy is to leverage known peptide-target interactions, as peptide ligands provide rich interface-level constraints. However, a major algorithmic challenge is that rigid atom-level conditioning on peptide structures can over-constrain the generative space and exclude valid small-molecule solutions. To address this challenge, we present PIFT (Peptide InterFace Transfer), a novel framework that transfers peptide-target interaction patterns to small molecules through joint sampling-time guidance of masked discrete diffusion and continuous flow matching. PIFT first maps peptide-target and small-molecule-target interfaces into a shared, target-conditioned interaction space, capturing transferable interface information across molecular modalities. The resulting peptide-derived reference then jointly steers building-block logits and 3D coordinates toward peptide-compatible molecules while preserving a strong adapted molecular prior. Extensive experiments across diverse PPI targets show that the learned representation captures cross-modal interaction compatibility, while peptide guidance improves interface recovery and achieves strong docking performance and favorable molecular properties.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41322
PDF
-
From 1D Benchmarks to Complex 3D Geometries: Can Agents Reliably Specify Problems for Automated PDE Simulation?
2026-10-30 05:20:28Akshat Jani, Sakhinana Sagar Srinivas, Venkataramana RunkanaAbstract
Turning an engineering request into a simulation requires consistent geometry, physics, boundary conditions, and numerical settings. Recent agent-assisted systems automate simulation setup, but successful execution alone does not establish whether the specification preserves the intended problem or correctly handles missing information. We study whether agents can reliably construct complete executable JSON specifications for automated geometry creation, meshing, PDE solving, and result export. Our workflow combines model extraction with deterministic scientific and capability checks, explicit clarification, and validated local patches that preserve unrelated accepted fields. We evaluate specification correctness, execution completion, and numerical verification separately. On 120 template-related requests, the full workflow achieves 98.3% correct first-turn specification and decision outcomes, compared with 92.5% for schema-assisted single shot and 82.5% for model-only: gains of 5.8 and 15.8 percentage points, respectively. It makes no false-ready decisions among the 42 requests requiring clarification or rejection. Component ablations identify contributions from scientific checks and conditional repair. Separate downstream studies cover 1D, 2D, and 3D steady and unsteady, linear and nonlinear problems, including flow and elasticity with complex three-dimensional wing geometries. A 30-case gallery demonstrates automated execution from retained JSON specifications, while 9 of 12 selected live user-to-results tests achieve joint specification, execution, export, and numerical-reference success (75.0%). These results support traceable, engineer-reviewed simulation assistance within the tested scope, while fresh-response regressions identify consistency as a priority for further improvement.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41267
PDF
-
Coarse3D: Controllable and Faithful Sparse Structure Generation for Image-to-3D
2026-10-30 04:09:19Gaige Wang, Bowen Pan, Hao Ren, Jiangjing Lyu, Chengfei LyuAbstract
Coarse-to-fine 3D-native generation relies heavily on the first-stage sparse structure, which determines the global orientation and overall geometry for subsequent refinement. However, existing coarse generators suffer from two key limitations: their orientation is difficult to control due to ambiguous semantic canonicalization, and their structures are often inconsistent with input images due to weak 2D-3D grounding. We propose Coarse3D, a sparse-structure generation model that produces orientation-controllable and structurally faithful voxel geometry from arbitrary unposed images. Coarse3D introduces three designs: a first-frame anchored canonical space that aligns generation with the first input view while preserving object-centric normalization; a dual-tower DiT that couples generation with auxiliary feed-forward reconstruction to learn spatially grounded features; and spatial-prior guided attention that regularizes cross-attention with pixel-to-voxel correspondence. Coarse3D can directly replace the coarse stage of existing pipelines. Experiments show that it substantially improves sparse structure quality and consistently enhances final 3D generation, achieving state-of-the-art performance.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41231
PDF
-
FlowMoDE: Coarse-to-fine Flow Matching for Detail-aware Sim-to-Real Monocular Depth Estimation
2026-10-30 03:13:53Liangjing Shao, Wanhao Liu, Jinsong Lin, Zhiwei Fang, Hongliang RenAbstract
Monocular depth estimation is a fundamental capability for 3D spatial perception in robotic and intelligent systems. While data-driven discriminative methods have achieved compelling performance for monocular depth estimation, generative approaches can reconstruct fine structural details with stronger generalization. However, most existing generative diffusion models suffer from low inference efficiency caused by multi-step sampling, whereas single-step diffusion models struggle to recover high-fidelity depth maps. To mitigate these drawbacks, this paper presents FlowMoDE, a coarse-to-fine flow-matching-based framework for efficient, structure-aware monocular depth estimation supporting sim-to-real and cross-scene generalization. The core architecture first conducts coarse-level decoding conditioned on diffused features extracted from a pre-trained semantic encoder, and subsequently restores fine-grained structural patterns via expanded decoding layers. Trained exclusively on a single synthetic indoor dataset, our proposed method outperforms prior methods across five real-world indoor and outdoor datasets, yielding an average 14.1% reduction in relative error against state-of-the-art works as well as 9.9\% reduction in chamfer distance for point cloud reconstruction. We further validate the generalization capability of FlowMoDE on realistic robotic scenes, with qualitative comparisons against state-of-the-art methods. The project page is this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.41190
PDF
-
MotorSense: A Video-EMG Dataset of Motoric Representations for Action Understanding
2026-10-30 01:32:19Eadom Dessalene, Michael Maynord, Amir-Hossein Shahidzadeh, Botao He, Yasamin Tabatabaee, Ruoshi Liu, Ruohan Gao, Paul Pu Liang, Yiannis AloimonosAbstract
Computer vision models learn hand action only from its visual consequences, without access to the underlying embodied motor signals. We address this absence with MotorSense, the first large-scale paired egocentric video and surface electromyography (EMG) dataset. MotorSense pairs bimanual EMG bands with egocentric video and yields synchronized video, bimanual EMG and IMU, and audio across 100 hours of unscripted manipulation. EMG offers a powerful scalable alternative to manual annotation: it captures finger movement, hand configuration, and the forces exerted between hand and environment as a dense signal free of occlusion and labeler bias, whereas manual annotation is costly, subjective, and unreliable under occlusion. We use MotorSense to learn sensorimotor alignment, producing an organization of action that is shared across the visual and motoric spaces, through cross-modal pre-training between video and EMG. The resultant vision models are then transferred to downstream tasks that benefit from sensorimotor alignment: in this case, action recognition, 3D hand reconstruction, and 3D contact reconstruction. We further demonstrate a generative application in which EMG motor signals are used to condition a video diffusion model, yielding videos of fine-grained hand manipulation. Across four action-centric computer vision tasks and eight popular video datasets, we demonstrate that sensorimotor alignment provides a stronger foundation for visual representation learning than conventional self-supervised learning methods over video, outperforming every baseline across all tasks and datasets evaluated.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41127
PDF
-
Gaussian Splatting-based Volumetric Video Compression with Sparse 4D Anchors
2026-10-30 00:41:25Ge Gao, Siyue Teng, Chanqgi Wang, Fan Zhang, Nantheera Anantrasirichai, Jui Chiu Chiang, Wen-Hsiao Peng, David BullAbstract
Immersive video communication requires photorealistic, compact dynamic scene representations that support efficient rendering. 3D Gaussian Splatting (3DGS) is a promising approach, but dynamic 3DGS remains difficult to compress due to its dense primitives and spatiotemporal redundancy. Anchor-based formulations improve compactness through sparse scaffolds that share geometry and appearance across primitives. However, existing designs typically deform a single canonical scaffold and condition each primitive solely on its associated anchor, limiting their ability to model non-local dynamics and disocclusions and exploit inter-anchor correlations. To address these limitations, we propose SAGA, a volumetric video codec built on Sparse Anchor-assisted GAussian splatting representations. SAGA represents a dynamic scene using hierarchical sparse anchors in 4D space-time. Coordinate-based INR decoders generate fine anchors and Gaussian primitives from features obtained through kernel interpolation among nearby anchors, enabling parameter sharing across spatiotemporal structures. A fixed-size memory with a novel orthogonality-informed slot-update strategy further captures long-range dependencies among unstructured anchors for entropy coding. Empirically, \name achieves PSNR-based BD-rate reductions of 90.88%, 64.30% and 65.14% relative to GIFStream, TED-4DGS, and OMG4, respectively, on the Neu3D testset.
Abstract (translated)
URL
https://arxiv.org/abs/2609.33969
PDF
-
Segment Anything Pancreas: A Unified MRI Segmentation Model for Pancreatic Cancer Radiotherapy
2026-10-29 23:18:16Zach Eidex, Mojtaba Safari, Kundan Thind, Parag J. Parikh, Zhen Tian, Xiaofeng YangAbstract
Background: Limited tumor annotations and incomplete labels across MRI datasets hinder unified segmentation of pancreatic tumors and organs at risk (OARs) for MRI-guided radiotherapy. Purpose: We propose Pancreas-MedNeXt, a unified model that segments the whole pancreas, its subregions, tumors, and surrounding OARs from one MRI volume. Methods: The model uses a shared MedNeXt encoder-decoder with supervised whole-pancreas and subregion pretraining, followed by multitask fine-tuning. Three output heads predict the pancreatic subregions, tumor, and OARs (stomach, duodenum, kidneys). An annotation-aware loss allows training on partially labeled data without treating unannotated structures as background. Pancreas and tumor were evaluated by three-fold cross-validation on 142 PANTHER patients (92 diagnostic, 50 MR-Linac), subregions and OARs on Cyst-X and AMOS test sets, and pancreas and OARs on 33 external T1-weighted TotalSegmentator scans. Comparisons with nnU-Net, MedNeXt, VISTA3D, and Swin UNETR used the Dice similarity coefficient (DSC) with paired bootstrap 95% CIs. Results: Pancreas-MedNeXt achieved the highest mean pancreas and tumor DSC on diagnostic and MR-Linac MRI. Tumor DSC was 58.41% and 42.55%, exceeding task-specific MedNeXt by 4.94 (95% CI: 0.28-9.45) and 5.89 (95% CI: 2.37-9.72) percentage points, respectively. Whole-pancreas DSC was 85.73% and 70.66%. Pretraining improved head, body, and tail DSC by 2.01, 3.00, and 3.78 percentage points over training from scratch. It also achieved the highest DSC for the pancreas, duodenum, and right kidney on AMOS, and for all five organs on external MRI, exceeding the next-best method by 4.92-8.85 percentage this http URL: A single Pancreas-MedNeXt model segments tumors, pancreatic subregions, and OARs on diagnostic and MR-Linac MRI, which could streamline contouring for pancreatic cancer radiotherapy.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41063
PDF
-
Prior-Guided Single-Exposure HDR Novel View Synthesis via Base-Gain Representation
2026-10-29 21:44:16Armin Shafiee Sarvestani, Sheyang Tang, Jialu Xu, Zhou WangAbstract
Single-exposure high dynamic range (HDR) novel view synthesis (NVS) recovers HDR scenes from low dynamic range (LDR) views captured at a single exposure and estimates camera response function (CRF) for LDR re-exposure. Without HDR ground-truth, using single-exposure LDR views as the sole source of direct supervision underconstrains HDR reconstruction and cannot ensure its faithful recovery, motivating the need for learned LDR-to-HDR reconstruction priors. Moreover, single-exposure LDR views supervise CRF only over a limited range, necessitating a learnable prior for reliable re-exposure beyond it. We propose PriorGS, a 3D Gaussian Splatting framework that addresses both challenges by (1) guiding HDR recovery with pseudo-HDR supervision from 2D single-exposure HDR reconstruction methods while anchoring it to input LDR views through a closed reconstruction loop; and (2) regularizing CRF modeling with a learnable prior that guarantees monotonicity and promotes plausible responses. Yet cross-view inconsistencies in independently generated pseudo-HDR targets impair scene reconstruction when HDR is encoded directly as Gaussian's color attribute. We therefore introduce a 3D base-gain HDR representation that factors HDR radiance into an LDR base and an overlaid 3D gain field, promoting robustness to such inconsistencies and enabling 3D HDR appearance editing. Extensive experiments demonstrate PriorGS's substantial superiority over baselines.
Abstract (translated)
URL
https://arxiv.org/abs/2610.41011
PDF
