Overview

3D foundation models aim to represent the shape, structure, and meaning of objects and scenes in forms that transfer across tasks such as reconstruction, segmentation, robotics, and generation. Common approaches pretrain on point clouds, multi-view images, meshes, or medical volumes using self-supervision, masked prediction, and alignment with images or language; models are then adapted with fine-tuning, lightweight prompts, or distillation. Generative models also use these representations to create or edit 3D assets and scenes. A central challenge is balancing broad transfer with reliable performance across different sensors, domains, and scales, while improving efficiency and evaluating geometric consistency and robustness.

Latest papers

71 papers7 papers in the last four weeks, down 36% on the four weeks before. 0.0% of all new papers.

Nov 9, 2026

A forensic evaluation framework for egocentric foundation models: an EFM3D case study for crime scene reconstruction

Foundation models increasingly combine geometry, semantics, and spatial context for three-dimensional scene understanding, but their suitability for forensic crime scene reconstruction remains unclear. This study presents a structured forensic evaluation framework for egocentric foundation models and demonstrates it using the Egocentric Foundation Model for 3D (EFM3D). Four simulated indoor crime scenes were recorded with Meta Project Aria Gen~1 at Robert Gordon University's East Lodge House forensic training facility. The framework evaluates room-scale reconstruction, semantic scene understanding, computational performance, and forensic suitability. Across the four scenes, EFM3D generated meshes averaging 12,668±2,43012{,}668\pm2{,}430 vertices and 24,838±4,90324{,}838\pm4{,}903 faces, with 39,529 recorded semantic predictions spanning 21--24 categories at a mean confidence of 0.40. Because reference meshes and expert-labelled 3D ground truth were unavailable, these statistics are treated as descriptive outputs rather than measures of geometric or detection accuracy. Final inference runs processed 1,388±2031{,}388\pm203 snippets per scene at 0.69±0.010.69\pm0.01 samples/s and required 50.77±6.1250.77\pm6.12 minutes per recording. A seven-criterion capability profile indicated stronger support for room geometry, furniture localisation, scene context, and spatial relationships than for fine-scale forensic evidence. The results position EFM3D as a complementary tool for contextual documentation and scene review rather than a replacement for high-fidelity forensic capture. The framework provides a reproducible basis for evaluating future egocentric foundation models in specialised forensic settings.
Nov 8, 2026

Map2SLAM: From Streaming Map Prediction to Real-Time SLAM

Streaming 3D foundation models (3DFMs) estimate camera poses and dense geometry online, but they fall short of SLAM: their predictions are fixed once emitted, so loop closures cannot correct accumulated drift. With such models already providing strong geometric capability, a central question for SLAM becomes how to schedule their inference and how to keep its results as revisable state. We present Map2SLAM, which keeps LingBot-Map frozen, trains only a lightweight retrieval head, and builds SLAM on the model's inference interface, turning its predictions into observations of an external persistent SLAM state. The front-end tracks every incoming frame with the model's causal streaming inference and decides which frames are written to the persistent state as keyframes. The back-end asynchronously decides when and which past frames to re-observe, re-running the model's joint inference over them with fresh context; verified loop observations correct the trajectory online, and segment observations refine dense geometry after the sequence. All invocations share a single frame encoder pass per image, whose cached features also drive loop retrieval, so frequent re-invocation incurs no additional encoding cost. Map2SLAM achieves state-of-the-art performance on KITTI, VBR, and Oxford Spires, reducing mean ATE by over 40% compared with the best uncalibrated baselines, while maintaining real-time tracking at over 28 FPS at a resolution of 518×154518\times154 on an NVIDIA RTX 5090D GPU. Please visit our project page at https://robbyant-research.github.io/Map2SLAM/.
Oct 24, 2026

ProMo3D: Probing Motion Cues in Frozen 3D Foundation Models

Feed-forward 3D foundation models have recently reshaped multi-view reconstruction, yet dynamic 4D understanding still often relies on large-scale motion supervision or hand-crafted, architecture-specific heuristics. We ask whether frozen 3D foundation model tokens already encode motion-relevant information that can be exposed by lightweight readout functions, rather than re-learning motion-specific features with high-capacity decoders. Through probing on three representative backbones, we find that static--dynamic separation is decodable from frozen token representations. Our rank-restricted and spectral analyses on π3π^3X reveal that the recovered motion cues are concentrated in a low-dimensional subspace. Building on this, we propose Probing-based Motion Distillation (PMD), which converts pairwise motion-state compatibility scores against a fixed anchor token into dense motion masks. Trained on only 160 YouTube-VOS videos using a single 16 GB GPU, PMD transfers zero-shot to DAVIS, SegTrackv2, and FBMS-59 without target-benchmark fine-tuning. With multi-view pretrained geometric backbones, PMD outperforms training-free heuristics and approaches substantially heavier supervised and self-supervised methods, suggesting that motion decoding from frozen 3D foundation models can be lightweight and data-efficient.
Oct 21, 2026

Plausible but Unimplantable: A Trustworthiness Audit of Shape Completion Foundation Models for Cranial Defect Reconstruction

A patient who has lost part of the skull vault needs an implant that fits the defect exactly, and such implants are still drawn by hand in CAD by specialists, which makes each case slow and expensive. The task can be considered as a shape completion problem, so pretrained completion models, and now the first shape completion foundation models, look like a way to remove that manual step. We audited three publicly released pretrained shape completion models, two of them foundation models, applied zero-shot to 239 cranial defects, some carved from public skull datasets and some real post-craniectomy cases. Each output was compared with simply returning the patient's own remaining bone, and tested for whether the predicted material could be placed without intersecting the bone that is still there. On the 210 carved defects none of the three improved on doing nothing, and on the 29 real cases only one did so beyond chance. Yet not one output, from any model in any configuration, was implantable: none managed both to stay clear of the bone the patient still has and to place more material than doing nothing, where a task-specific volumetric model, trained for this task and selected on these cases, was implantable on all but one. The output that most convincingly resembles a skull is no more placeable than the rest, and at these point densities implant thickness cannot be read off the emitted points at all, which is a property of the output representation rather than of any one model. On this evidence the shape completion foundation models we could obtain cannot be trusted and, used as released, are not usable in clinical practice. Substantial research towards trustworthy foundation models for shape completion is still needed.
Oct 16, 2026

Foundation-Model-Driven 3D Scene Graphs for Mobile Manipulation

Geometric and Semantic Foundation models are becoming increasingly popular tools for generating spatial maps for robots. Recent work (e.g., FOUND-IT) has leveraged these tools to construct real-time, task-driven, and open-set 3D scene graphs from a single uncalibrated camera. However, while classical robot mapping techniques have undergone extensive system demonstrations, limited work has been done to develop both the fundamental and the systems wide advancements needed to integrate these new types of robot spatial memory into a full, useable autonomy pipeline. In this paper, we bridge this gap, and in doing so we also evaluate the effectiveness and limitations of this type of 3D memory. Towards this end, we integrate FOUND-IT into a mobile manipulator platform and demonstrate it is well-suited to support multiple objectives such as object change detection, support for multiple mapping sessions, determination of navigable paths, and directly support object grasping for the task of scene re-arrangement. Furthermore, we provide nuanced insights into the effectiveness and limitations of a monocular, foundation-model-driven 3D scene graph construction system using observations from rigorous on-robot experiments. Code is available at https://github. com/MIT-SPARK/FOUND-IT.
Oct 15, 2026

Towards Reconstructing Geographically Diverse Architecture with 3D Foundation Models

Internet-scale photo collections of real-world landmarks have driven progress in 3D computer vision yet remain highly challenging for modern 3D foundation models (3DFMs) that estimate scene structure in a single feed-forward pass. In this work, we introduce ArchWorld, a benchmark of 3D architectural landmarks with high geographic coverage and rich metadata. We use this benchmark to conduct a thorough error analysis of current 3DFMs to understand where they fall short and how they can be improved. We examine socially relevant disparities in model error and find that model performance varies by geographic region; however, unlike typical fairness regimes, the trends only partially match training data representation. Our investigation reveals that the observed geographic disparities reflect not only differences in data representation and scene size but also differences in region-specific architectural features and background distractors that make scenes from different regions inherently vary in difficulty. Leveraging insights from these model failings, we introduce an intervention scheme that improves 3DFM performance while simultaneously reducing geographic disparities.
Oct 14, 2026

Remember, Revisit, Revise: Scaffold Memory for Long-Horizon 3D Geometry

Geometry foundation models estimate camera poses and 3D scene structure from video sequences, but maintaining consistent reconstructions over long horizons remains challenging. Feed-forward models process long sequences as independent, fixed-length frame windows and align their predictions, causing pose and reconstruction errors to accumulate across windows. Streaming models retain information from past frames to inform future predictions, but cannot revise earlier estimates when later observations reveal errors. To address this, we introduce Scaffold MemOry for Revision (SMORE), a framework equipping frozen geometry foundation models with a revisable external memory for long-horizon inference. Inspired by hippocampal associative memory in biological brains, which links spatial context with sensory experience, SMORE associates where an observer is with what it observes, enabling past observations to be stored, retrieved, verified against new evidence, and revised when inconsistencies are detected. We evaluate SMORE with eight competitive geometry models on five datasets spanning camera pose estimation, multi-view depth estimation, 3D point-cloud reconstruction, simultaneous localization and mapping (SLAM), and novel view synthesis. Across all evaluated models, SMORE consistently improves long-sequence pose estimation, reduces multi-view depth error by up to 58%, reduces Chamfer distance for 3D point-cloud reconstruction by up to 43%, and achieves average trajectory error as low as 2.9 cm in long-horizon SLAM. By retaining corrected poses and observations, SMORE further improves the fidelity and spatial consistency of large-scale 3D scene reconstruction. These results establish revisable external memory as a flexible interface for extending pretrained geometry models beyond fixed-context prediction toward iterative, error-correcting inference over long visual sequences.
Oct 12, 2026

UniMetric: Universal Metric Scale Recovery for Visual Geometry Foundation Models

Recent advances in multimodal foundation models, exemplified by GPT-6 Astra, raise the prospect of general-purpose embodied reasoning and planning. Using Internet videos for embodied pretraining calls for reliable physical-scale annotations. Yet visual geometry foundation models often recover scene structure and camera motion without reliable metric scale. We show that uncertain size priors of familiar objects can jointly constrain a reconstruction's shared scale. Building on this insight, we present UniMetric, a universal and training-free metric scale recovery framework for geometry foundation models, without additional sensors or task-specific metric supervision. Specifically, Metric Prior Acquisition elicits object height intervals from a vision-language model. Semantic-Geometric Alignment measures reconstructed object heights along the estimated support-surface normal in each frame and converts the priors into scale intervals. Global Scale Recovery reconciles these intervals through a convex objective, yielding a jointly feasible scale when they overlap and a unique compromise otherwise. The estimated sequence-level scale is applied uniformly to reconstructed points and camera translations. Controlled experiments with real VLM priors show that combining observations improves scale accuracy with exact relative object heights. To evaluate metric accuracy on GFM reconstructions without ground-truth scale alignment, we introduce MetricBench, covering twelve embodied, driving, and roadside subsets with sensor-derived ground truth. Across nine geometry foundation models, UniMetric reduces mean depth Abs Rel, reconstruction Chamfer distance, and trajectory error by 54.2%, 83.8%, and 65.6%, respectively, improving every model on all three tasks at the suite level. We release the UniMetric code and MetricBench data.
Oct 10, 2026

Drive3R: Block-wise Streaming Geometric Foundation Models for Autonomous Driving

High-fidelity closed-loop simulation for autonomous driving increasingly relies on dense multi-view 3D reconstruction. While recent feed-forward geometric foundation models have revolutionized per-scene optimization, applying them to continuous, long-horizon driving sequences is computationally intractable due to GPU memory limits and the quadratic complexity of global attention. Existing scaling strategies are fundamentally limited, as causal streaming suffers from compounding scale drift and trajectory deviation, while rigid "chunk-then-align" heuristics fragment global context and produce trajectory discontinuities. To address these limitations, we introduce Drive3R, a block-wise streaming geometric foundation model. Its block-wise causal attention resolves the scalability bottleneck of fully bidirectional attention while avoiding the loss of local future context in purely causal streaming. Its confidence-aware re-anchoring further mitigates long-range drift by detecting unreliable streaming states, re-initializing the KV cache, and stitching refreshed segments through SE(3) alignment. Extensive experiments demonstrate that Drive3R establishes a new state-of-the-art on nuScenes, KITTI, and Waymo. Specifically, on nuScenes, it improves over the recent DVGT-2 baseline from 0.1806 to 0.1081 in AbsRel, from 0.6750 to 0.5344 in point-cloud Accuracy, and from 1.5812 to 0.9092 in ATE.
Oct 8, 2026

Zebra3D: Efficient Hybrid Foundation Models For Scalable 3D Reconstruction

Feed-forward 3D foundation models reconstruct scenes from unposed images, but the quadratic cost of global attention limits their scalability to large view sets. We present Zebra3D, a layerwise sparse-linear hybrid attention framework that improves the efficiency of pretrained 3D foundation models without training from scratch. Across VGGT, HunyuanWorld-Mirror, and π3π^3, we observe a broadly consistent depth-wise pattern: early stages are relatively tolerant to linear replacement, while sensitive middle-to-late stages tend to exhibit more concentrated attention. Zebra3D exploits this structure by assigning linear-centered attention to tolerant stages and tile-sparse attention to sensitive stages, using landmark summaries to preserve unselected context. Closed-form affine fitting enables training-free conversion, while optional post-conversion training combines task supervision with output and relational feature distillation. Across all three backbones, Zebra3D achieves strong reconstruction and camera-pose accuracy at high attention sparsity, with training-free variants remaining close to their dense counterparts. End-to-end speedups increase with input view count, reaching up to 7.1×7.1\times at 1,024 views. After post-conversion training, Zebra3D-VGGT reaches 80.5% RealEstate10K AUC@30 at 94% mean attention sparsity, within 1.1 percentage points of dense VGGT fine-tuned on the same data and schedule.
Oct 8, 2026

Pumpire: Unified Benchmark for Metric Distance Estimation

We present Pumpire, a unified benchmark for evaluating metric point-pair distance estimation capability of both image- and video-level 3D foundation models, with or without depth priors. In contrast to previous approaches that normally evaluate depth and camera intrinsics separately or evaluate point-clouds with geometric similarity metrics, which cannot directly reflect models' point-to-point distance estimation capability, Pumpire directly assesses point-to-point distances from the reconstructed geometry. To this end, we collect a large-scale and diverse dataset (pumpire-6k) comprising 100 real-world scenes, each annotated with physically measured point-pair distances and containing 64 frames, for a total of 6,400 frames. Building on this dataset, we establish a holistic evaluation protocol that covers both image- and video-level 3D foundation models and enables direct assessment of point-pair distance errors and cross-setting comparison. We conduct extensive experiments across 29 baseline configurations of representative 3D foundation models and provide a comprehensive analysis of the results. By offering this benchmark, we target the more fundamental ability to perceive and estimate physical scale in the reconstructed 3D space, which prior evaluation protocols have largely overlooked. The project page can be found at https://pumpire.github.io/
Oct 6, 2026

RDGSplat: Render-Dedicated Geometry for Novel View Synthesis

Recent 3D foundation models extend to novel view synthesis by carrying a Gaussian head on the representation they already use for reconstruction. However, their rendering quality lags far behind their reconstruction accuracy, because the renderer reads a geometry and a representation that are both optimized for metric reconstruction. Prior methods alleviate this by updating the backbone weights, but they thereby sacrifice the metric predictions the model was built for. To this end, we propose RDGSplat, which decodes a geometry dedicated to rendering from a frozen 3D foundation model while keeping its metric predictions exact. In particular, we devise Render-Dedicated Geometry Decoding, which duplicates the pretrained decoders and optimizes the duplicates under photometric supervision alone. Then, a Target-Pose Conditioned Adapter is introduced to adapt the representation those decoders read towards rendering, conditioned on the predicted target pose. Extensive experiments show that RDGSplat improves novel view synthesis for three feed-forward models of two architectures on four benchmarks, with every pretrained weight frozen. On RE10K, it raises WM2.0 from 20.918 to 25.149dB by training 205.5M added parameters against a frozen 1.4B backbone, and leaves its predicted depth and pose unchanged.
Oct 5, 2026

Less Context, Better Geometry: Masked Geometric Encoder for Robust 3D Foundation Models

Recent progress in 3D foundation models has enabled rapid 3D reconstruction and camera calibration by leveraging learned 3D priors from vast amount of spatial data. However, the all-to-all global attention design leads to quadratic complexity and limits long-sequence inference; unconstrained cross-view interactions also can propagate unreliable evidence from occluded or visually similar but geometrically distant views. In this paper, We introduce a Masked Geometric Encoder (MGE), which promotes the learning of robust geometric representations under incomplete cross-view context. During training, MGE strategically drops frame tokens from global attention and distills from a pretrained full-context teacher model. This allows the model to learn an intrinsically richer per-frame representation while providing sufficient intermediate supervision to avoid performance degradation. Through extensive experiments, we show that MGE leads to much stronger performance under occlusion and doppelganger views while retaining high performance on standard benchmarks. Such a richer frame representation also leads to more effective token reduction during inference. To this end, we develop a novel Anchor-Guided Adaptive token merging technique that preserves representative anchor frames while jointly merging redundant tokens from the remaining views. Compared to other efficient inference approaches, we can achieve inference speedup while consistently maintaining higher reconstruction quality, particularly in limited-view settings.
Oct 1, 2026

CLoSeR: Closing the Loop for Long-Context Streaming Reconstruction

Feedforward foundation models have recently shown remarkable 3D reconstruction capabilities. However, existing models exhibit large tracking drift in long-context streaming reconstruction due to error accumulation. In this paper, we revisit loop closure with streaming reconstruction foundation models to enable accurate, drift-free, kilometer-scale reconstruction. Specifically, our method detects loop candidates through global descriptor retrieval, and constructs loop-conditioned windows to estimate the relative poses between looped frames. Given the observation that our adopted streaming reconstruction backbone produces a globally consistent scale, we optimize all frame poses on the SE(3) manifold with sequential and loop closure constraints, avoiding the pose graph optimization on the Sim(3) or higher-dimensional SL(4) manifolds employed in prior works. Extensive experiments show that our method reduces drift and produces consistent geometry on kilometer-scale sequences, significantly outperforming the state of the art. Code is available at https://github.com/MoyangLi00/CLoSeR.git.
Sep 29, 2026

TaoFlowForge: Progressive Native Mesh Generation via Cascaded Flow Matching

3D content generation technology has significantly advanced the work of designers, as well as the 3D printing and gaming industries. However, it remains difficult to produce lightweight, editable, and topologically clean artistic content that is directly production-ready. To achieve this, we present TaoFlowForge, an artistic mesh foundation model that generates production-ready meshes. Specifically, TaoFlowForge decomposes the mesh generation process into vertices generation and their connectivity prediction, i.e., edges. We formulate vertices generation as a two-stage coarse-to-fine process and incorporate several effective loss functions to further enhance its performance. In the connectivity prediction stage, we propose a simple yet effective method for estimating the connectivity affinity between vertices and additionally predict per-vertex normals, which determines the correct orientation of faces. Besides, we construct a large-scale dataset combining hand-crafted 3D assets with public high-quality topology datasets. Based on this, a carefully designed data curation pipeline is employed to filter the raw dataset, retaining only high-quality topology data for model training. Our model is trained on the combined dataset and tested on both out-of-distribution hand-crafted set of 3D assets and public datasets. Under image-conditioned generation, TaoFlowForge outperforms autoregressive methods and achieves state-of-the-art results among open-source mesh topology generators. We will release all the code and weights together with a portion of our test dataset.
Sep 22, 2026

nnFoundation: 3D Foundation Models for Radiology

Radiological artificial intelligence has advanced rapidly, yet most systems remain narrowly task-specific, data-intensive, and fragile under domain shift. Foundation models promise more transferable and data-efficient solutions, but existing approaches are limited in scale, evaluated narrowly, and often assume that a single pretrained model can support diverse downstream tasks. Here we present nnFoundation, complementary convolutional and transformer-based 3D radiological foundation models. Developed within the Human Radiome Project (THRP), nnFoundation is trained on 2.1 million CT, MRI, and PET image volumes from 125 institutional and public datasets. We evaluate them across 108 tasks spanning segmentation, detection, classification, report generation, and image retrieval, including evaluations under domain shift, by external partners and in low-data and low-compute regimes. Across all task types, our convolution- and transformer-based nnFoundation models consistently outperform both prior 3D foundation models and training from scratch, establishing state-of-the-art performance for radiological imaging. However, performance follows a consistent task-dependent structure: the convolutional nnFoundation model dominates spatially localized tasks, whereas the transformer-based nnFoundation model excels in tasks requiring global semantic reasoning and in frozen-feature settings. Dynamically aligning the foundation model topology with the dataset characteristics post-hoc further improves transfer across heterogeneous 3D settings. These results show that transferable 3D radiological performance is governed not by a single universal model, but by the interplay of scalable pretraining, complementary architectures, and dataset-aware adaptation. We release nnFoundation models integrated into nnU-Net and nnDetection, enabling immediate application across established radiology workflows.
Sep 21, 2026

Toward a foundation model for forest point clouds

Forest inventories increasingly rely on artificial intelligence (AI) models to derive forest attributes from large-scale 3D point clouds. Current models are typically specialized to a single task, sensor, and forest type, making adaptation expensive in terms of annotations, computation, and expertise. We ask whether a single pretrained model can instead learn transferable representations across diverse forest inventory settings. Inspired by recent developments in language modelling and computer vision, we take a step toward a foundation model (FM) for 3D forestry. Using LitePT as backbone, we first establish a strong supervised baseline that sets a new state of the art on forest semantic and instance segmentation, tree species classification, and age regression benchmarks. We then curate a large-scale unlabelled corpus spanning airborne, UAV, and mobile laser scanning across diverse forest ecosystems, and pretrain the same backbone using self-supervised learning. We systematically evaluate representation learning strategies by comparing training from scratch, supervised pretraining, and self-supervised pretraining across four representative forestry tasks, under varying annotation budgets. Compared with training from scratch, self-supervised pretraining accelerates model convergence and consistently improves performance when annotations are scarce. Compared with task-specific supervised pretraining, self-supervised pretraining yields more transferable representations across downstream forestry tasks. These findings identify the practical regime in which pretrained representations are most valuable and suggest that instance discrimination, rather than forest semantics, is the main remaining obstacle to a general-purpose 3D forest foundation model. Code and models are available at: https://github.com/prs-eth/ForPT.
Sep 18, 2026

Adaptive World Memory 3D Foundation Model for Scalable 3D Mapping, Localization, and Rendering

Recent 3D foundation models enable generalizable geometric reasoning from RGB images but remain limited in persistent memory, scalability, and renderable scene modeling. We present a memory-centric 3D foundation model for scalable robotic localization, reconstruction, and Gaussian rendering. Its core is an adaptive world memory mechanism that combines transformer-based gated updates with test-time temporal-spatial regulation. Learned gates control recurrent memory propagation, while temporal state evolution and spatial observation-state consistency regulate token-wise updates and forgetting over long image sequences. To support large-scale mapping, we organize memory into local submaps and integrate progressive mapping and tracking, loop closure, and SL(4)-based global refinement to maintain local accuracy and global consistency. A Gaussian reconstruction head decodes memory-enhanced features into renderable primitives, unifying camera pose estimation, dense point-cloud reconstruction, and photorealistic rendering within a single model. Experiments on public benchmarks and self-collected datasets from diverse robotic platforms demonstrate improved trajectory accuracy, reconstruction completeness, and rendering quality over existing 3D foundation reconstruction and SLAM baselines. These results support adaptive memory as a foundation for persistent robotic world modeling. The dataset and code will be made publicly available at https://github.com/dtc111111/AWM-3DFM.
Sep 16, 2026

GeoCond: A Conditioning-Aware Reliability Adapter for Feed-Forward 3D Reconstruction

Feed-forward 3D foundation models such as VGGT predict cameras, depth, and point maps in a single pass, but can fail silently under low overlap, low parallax, and extreme relative rotation. Stratified analyses over these factors show that these failures are governed by geometric conditioning and are poorly captured by native aleatoric confidence. We introduce GeoCond, a lightweight reliability adapter for frozen feed-forward 3D backbones. GeoCond reads geometric conditioning cues and the backbone's own stability signals and outputs pose-level uncertainty and a refinement gate. During training, it can be supervised by frame-permutation orbit variance, ground-truth pose error when labels are available, or cycle residuals from unlabelled independent pose graphs. At inference, the default head requires only one backbone pass and a small MLP. On VGGT, GeoCond improves out-of-distribution (OOD) AUSE (area under the sparsification-error curve; lower is better) from 0.320.32 to 0.200.20 over native confidence on held-out outdoor scenes (and to 0.210.21 when trained on indoor data only) and avoids the collapse caused by applying bundle adjustment uniformly, without improving on feed-forward poses there. For three of six backbones, cycle-distilled variants provide a ground-truth-free adaptation route, including cases where permutation variance vanishes on equivariant models. The same reliability signal supports gated refinement, pose-graph weighting, calibration, curation, and capture decisions. Reliable feed-forward 3D reconstruction requires not only predicting geometry, but also knowing when that geometry should be trusted.
Sep 10, 2026

Beyond Argmax: A Mechanistic Study of Semantic Retention in Frozen Foundation-Model Composition for Generalized Few-Shot 3D Segmentation

Classical classifier-combination work distinguishes score-level fusion from hard decision-level voting. We revisit this distinction where independently pretrained, frozen foundation models are composed at inference time for generalized few-shot 3D segmentation. We ask: how much useful semantic information is lost when heterogeneous sources are collapsed to a single class before they can interact? We answer with a same-input semantic-retention intervention. Dense RegionPLC and sparse cross-view SAM3 evidence, model weights, masks, geometry, vocabularies, and fusion rules are frozen; only the number of semantic alternatives retained before interaction is varied via a matched top-k ladder. On 156 held-out ScanNet200 scenes, top-1 reaches 28.47 harmonic-mean (HM) IoU while full distribution fusion reaches 34.87 HM (+6.40, 95% CI [+5.24,+7.64]). The pattern replicates on 50 ScanNet++ scenes: 23.02 vs. 26.50 HM (+3.48, 95% CI [+1.64,+5.93]). The conclusion is robust: full-distribution HM is stable across sparse-source weights 0.3--0.7; alternative operators (max, geometric pooling) also outperform top-1; and a GroundingDINO--SAM2.1 source-replacement diagnostic shows monotonic HM increase from 14.77 to 18.75 with full retention. Calibration diagnostics reveal opposite miscalibration of the two sources, yet correcting calibration does not eliminate the retention advantage. Across datasets and source stacks, most information is recovered by retaining a compact set of plausible alternatives. The contribution is a controlled diagnosis of premature semantic collapse as a repeatable information bottleneck in heterogeneous frozen-model composition.
Sep 3, 2026

Zero-Shot Novel Depth Synthesis Using 3D Foundation Models Scene Representations

3D Foundation Models (3DFMs) such as VGGT have recently pushed the boundaries of 3D vision by predicting rich unified representations with feed-foward transformers. The scene representations learned by these models enable strong performance on multiple 3D vision tasks. In this paper, we investigate using their internal representations to infer 3D in the scene from new views. Our hypothesis is that in order to solve the task of 3D reconstruction, these models need to learn a representation that includes a large amount of general knowledge about 3D scenes. After showing that it is possible to decode hidden surfaces from internal 3DFM representations, we propose a method, Z3D, that estimates pointmaps in unseen views by doing latent diffusion on 3DFM representation. We show that Z3D can predict realistic depth maps for new views across multiple datasets.
Sep 3, 2026

VI3: Grounding Pretrained 3D Foundation Models with Inertial Cues

3D foundation models (3DFMs) excel at predicting camera poses and dense depth from multiple views of a scene, showcasing strong zero-shot generalization. However, as metric scale is not observable from monocular images, their absolute scale predictions are typically inaccurate. Inertial measurement units (IMUs), present in most devices, naturally complement monocular cameras by observing scaled motion. We introduce VI3, a model-agnostic framework that metrically anchors a pretrained 3DFM using only IMU readings. VI3 initializes and preintegrates the IMU to obtain a metric motion reference, which is then used to recover the scale of the 3DFM outputs. Our method includes adaptable anchoring strategies tailored to diverse 3DFM architectures. Experiments on synthetic and real aerial datasets demonstrate that VI3 recovers metric scale without ground-truth supervision while preserving geometric consistency, acting as a fine refinement under well-conditioned motion and as a strong prior when motion is less informative.
Sep 1, 2026

Cross-Model Distillation of a Human-Pose Foundation Model from Unannotated Infant Video for Markerless 3D Pose Estimation

Spontaneous movement is one of the earliest windows onto an infant's neuromotor health, and structured clinical instruments that score it are validated early predictors of cerebral-palsy risk. However, they require specially trained raters, are time-consuming, and carry inter-rater variability. This motivates automated, video-based markerless assessment, especially as marker-based motion capture is impractical in infants. Yet the foundation models that make markerless capture possible are trained almost entirely on adults: our recent multi-view infant study found that no single model is jointly best, with strong 2D keypoint accuracy and direct 3D body recovery split across different models. While that study identifies this trade-off, it does not resolve it. Here, we perform cross-model distillation from the Sapiens 2 pose model into the SAM 3D Body model, using unannotated infant video alone. A frozen teacher supplies dense pseudo-labels, and a differentiable renderer aligns the predicted mesh to them in the training loop. On eleven held-out infants (18 sessions, 173 recordings) under our prior study's multi-view protocol, fine-tuning improves same-view 2D keypoint agreement with the Sapiens reference (median body percentage of correct keypoints @ 10px 0.22 -> 0.42, face 0.22 -> 0.42) and Procrustes-aligned mean per joint 3D position error (25.5 -> 22.2 mm). This demonstrates how cross-model distillation improves SAM 3D Body model performance on infants.
Aug 30, 2026

GeoRay: Gauge-Aware Feed-Forward Satellite 3D Reconstruction in the Geodetic Frame

Feed-forward 3D foundation models reconstruct perspective scenes in one pass. Satellite photogrammetry needs a different product, one that domain adaptation alone does not deliver: dense surface height in an absolute geodetic frame under non-central rational polynomial cameras (RPCs). Perspective-pretrained features are not reliably observable along RPC height rays, absolute elevation carries a low-order height--datum gauge exchangeable with sensor bias to first order, and monocular and multi-view cues fail in different regions. \method{} treats all three. Lightweight ray-consistent adapters make a frozen backbone matchable along native RPC rays. An explicit datum mechanism separates relief from absolute level and is equivariant to the vertical origin by construction, so one trained model serves zero-, one-, and sparse-control inference. Calibrated inverse-variance fusion combines the two relief streams. \bench{}, our absolute-frame benchmark of eighteen systems across in-domain, cross-dataset, and cross-city tiers, scores absolute placement without registration or test-reference leakage. On 26 held-out US3D tiles, \method{} attains 2.992.99,m absolute MAE at 91.9%91.9\% coverage, improves completeness-aware accuracy by 46.446.4 points over the strongest compliant feed-forward baseline, remains the most accurate such system under both transfer shifts, and runs in 2424,s model-forward time per tile. Code and models will be released at https://github.com/HIT-SIRS/GeoRay
Aug 21, 2026

Lift, Associate, and Fuse: A Decision-Centric Framework for 2D-to-3D Foundation Model Transfer

Methods that transfer predictions from two-dimensional foundation models into three-dimensional segmentation are commonly grouped by task or representation. Those groupings obscure the decisions that determine whether a system remains coherent across views: where image evidence is grounded, when observations become one identity, how semantic and granularity conflicts are handled, which information is fused, and what state survives for later queries. We introduce Lift, Associate, and Fuse (LAF), a decision-centric framework that represents a transfer system as five operators: Generate, Associate, Reconcile, Fuse, and Persist/Query. LAF defines an explicit contract for the persistent carrier---its spatial support, semantic state, identity state, uncertainty, provenance, and supported operations---and identifies the first stage at which discarded evidence becomes unrecoverable. We operationalize the framework as a structured audit protocol and apply it to 161 systems available through 7 August 2026, spanning point-, field-, Gaussian-, object-, graph-, and memory-based carriers. Representation, temporal, relational, and feed-forward stress tests required no additional analytical stage after the final confirmation pass. The resulting decision traces expose four recurring properties: association does not establish identity; carrier design fixes both the query interface and correction boundary; rendered-view, native-3D, and proposal-level evaluations are not interchangeable; and qualifiers such as training-free, real-time, open-vocabulary, and generalizable are meaningful only when attached to a stage and a complete cost ledger. LAF therefore supplies a representation-neutral method for comparing existing systems, diagnosing irreversible failures, and specifying revisable 3D perception for future agents.
Aug 6, 2026

Do 3D Medical Foundation Models See Through MRI Artifacts? A Controlled Study of Representation Robustness

Self-supervised 3D medical foundation models are increasingly used as general-purpose feature extractors, yet their sensitivity to MRI artifacts remains poorly understood. We present a controlled evaluation of representation robustness across five pretrained 3D encoders spanning different architectures, objectives, pretraining domains, and dataset scales. Using BraTS-Africa cases with four MRI sequences, we generate seven frequency- and image-domain artifacts at five predefined corruption settings. Robustness is assessed using linear centered kernel alignment (CKA), RankMe, and UMAP, complemented by an independent segmentation-consistency analysis. We find that robustness is strongly model- and artifact-dependent. 3DINO exhibits the most consistently stable representations, while BrainIAC is highly sensitive to several corruptions; NeuroVFM, BrainFM, and Neuro-SimCLR show intermediate but distinct artifact-specific profiles. Across many conditions, CKA decreases substantially while RankMe remains comparatively stable, indicating that artifacts often distort representation geometry without causing dimensional collapse. Segmentation consistency also degrades under corruption, particularly for ghosting and Rician noise, but aligns only partially with representation-level robustness. These findings show that larger-scale or domain-specific pretraining alone does not guarantee artifact invariance and motivate explicit robustness evaluation before deploying 3D foundation models in heterogeneous MRI settings.
Aug 6, 2026

Big, Bright, or Invisible: A Frozen-Feature Benchmark of 3D CT Foundation Models

Routine CT interpretation is inherently comprehensive, capturing incidental findings across the entire scan volume. 3D CT foundation models could assist this process by providing generalizable representations of anatomy and pathology. To evaluate their diagnostic breadth, we benchmark ten frozen CT encoders across three cohorts of thoracic CT scans, including an unseen internal clinical dataset, using kk-nearest neighbors, zero-shot prompting, and linear probing. We find no universal state-of-the-art, with rankings fluctuating significantly depending on the evaluation context. While models combining fine-grained image tokenization with vision-language alignment generally perform best, a lightweight supervised encoder remains highly competitive, demonstrating that explicit labels can effectively substitute for scale. Crucially, rather than model architecture, we observe that the primary determinant of performance is a physical bottleneck: a finding's detectability scales with its contrast against surrounding tissue and its spatial extent. Through controlled within-organ comparisons, we empirically demonstrate that widespread or high-contrast abnormalities, such as devices and effusions, are reliably recovered. Conversely, small, low-contrast focal lesions remain a persistent challenge across all evaluated encoders. We attribute this to the inherent limitations of globally pooled embeddings, suggesting that accurately representing small, low-contrast structures will require region- or lesion-level pretraining.
Aug 4, 2026

Qwen-3D: A Generalist 3D Vision-Language Model for Spatial Understanding

Large Multimodal Models (LMMs) have achieved remarkable success on images and short videos, yet scaling them to long videos remains challenging due to frame-centric tokenization and limited context windows. 3D geometry provides a natural compression mechanism for visual streams: depth and camera pose enable observations from multiple views and time steps to be fused into a persistent, world-aligned representation. While recent 3D LMMs leverage geometry-aware representations to improve spatial reasoning, they continue to lag behind specialist 3D perception systems on grounding and segmentation tasks. We argue that a key limitation is geometry-aware decoding: existing methods communicate 3D predictions through language tokens, proposal selection, or lightweight grounding queries, creating a bottleneck between language reasoning and dense geometric prediction. Building on these insights, we introduce Qwen-3D, a geometry-aware LMM that compresses visual information within the Qwen backbone using multi-view geometric cues, enabling efficient long-horizon visual reasoning over static scenes. Qwen-3D augments visual tokens with 3D Rotary Positional Embeddings, allowing attention to operate directly in 3D scene space rather than across independent image frames and thereby facilitating scalable cross-view and temporal reasoning. To bridge language and geometry, Qwen-3D incorporates a query-based segmentation decoder that grounds language directly in the underlying 3D scene representation, unifying referential grounding, instance segmentation, and visual question answering across both images and videos. Across a diverse set of benchmarks, Qwen-3D surpasses existing 3D LMMs and outperforms several large proprietary 2D models. Notably, Qwen-3D achieves these improvements while maintaining strong performance on standard 2D vision-language benchmarks by jointly training on 2D and 3D data.
Aug 3, 2026

Hunyuan3D-Buffalo 1.0: A Unified Multimodal Model for Scalable 3D Generation, Understanding, and Editing

Recent advances in image generation have demonstrated the potential of unified multimodal models that integrate understanding, generation, and editing. However, unified 3D modeling remains constrained by scarce multimodal data, particularly the lack of large-scale and geometrically consistent editing data. To address this limitation, we propose Hunyuan3D-Buffalo 1.0, a unified framework supporting 3D understanding, text-to-3D generation, instruction-guided 3D editing, and text-grounded part generation within a single architecture. To enable scalable training, we construct an 87M-scale 3D multimodal corpus, comprising 25M understanding samples, 50M text-to-3D pairs, and 12M editing pairs generated using Nano3D-v2. Architecturally, the framework combines Hunyuan3D-VLM for semantic, structural, and spatial understanding with Hunyuan3D DiT for high-fidelity 3D synthesis. The VLM provides multimodal semantic conditions for generation, while editing and part generation additionally condition the diffusion process on the source object representation to preserve its overall structure and unedited regions. Extensive experiments show that Hunyuan3D-Buffalo 1.0 achieves state-of-the-art or leading performance on text-to-3D generation and 3D editing benchmarks, while exhibiting strong understanding and part-generation capabilities. Our analysis further shows that both generation and understanding improve editing, demonstrating the effectiveness of unified 3D multimodal training. Project Page: https://tencent-hunyuan.github.io/Hunyuan3D-Buffalo1.0/
Jul 27, 2026

UMI3D: Robust 3D Generation on Unconstrained Multi-Image Inputs via Simultaneous Focus Cross-Attention Routing

Recent 3D foundation models can generate high-quality assets from a single image, but degrade markedly on unconstrained multi-image inputs, often producing distorted geometry, over-smoothed textures, and chaotic colors. We argue that this failure stems not from limited model capacity, but from a mismatch between single-image cross-attention and the multi-image setting: existing models lack a principled way to decide which image each 3D voxel should trust at each denoising step. Revisiting recent single-image 3D foundation models, we show that explicitly routing each voxel to its most informative image is sufficient to unlock strong performance on inconsistent multi-image inputs. Based on this observation, we propose UMI3D, a training-free and plug-and-play framework that restructures cross-attention for unconstrained multi-image 3D generation. Its core, Simultaneous Focus Cross-Attention (SFC-Attn), activates all conditioning images at each denoising step while allowing each voxel to focus on the single image that best explains it. To enable this routing, we derive the Voxel Reference Score (VRS), a model-intrinsic metric for voxel--image affinity that requires no external matching, segmentation, or correspondence models. Extensive experiments show that UMI3D unlocks the multi-image potential of single-image 3D generation frameworks across diverse tasks. Project Page: UMI3D-Project.github.io.