SpatialBench evaluates 41 spatial foundation models across 6 paradigms and 5 task suites, finds they are not all-round players, and introduces the DA-Next-5M dataset plus DA-Next baseline model.
hub Tool reference
Stereo Magnification: Learning View Synthesis using Multiplane Images
Tool reference. 71% of classified Pith citations use this work as a method, library, or software dependency, not as a substantive claim.
53
Pith papers citing it
Method reference
71% of classified citations
abstract
The view synthesis problem--generating novel views of a scene from known imagery--has garnered recent attention due in part to compelling applications in virtual and augmented reality. In this paper, we explore an intriguing scenario for view synthesis: extrapolating views from imagery captured by narrow-baseline stereo cameras, including VR cameras and now-widespread dual-lens camera phones. We call this problem stereo magnification, and propose a learning framework that leverages a new layered representation that we call multiplane images (MPIs). Our method also uses a massive new data source for learning view extrapolation: online videos on YouTube. Using data mined from such videos, we train a deep network that predicts an MPI from an input stereo image pair. This inferred MPI can then be used to synthesize a range of novel views of the scene, including views that extrapolate significantly beyond the input baseline. We show that our method compares favorably with several recent view synthesis methods, and demonstrate applications in magnifying narrow-baseline stereo images.
hub tools
citation-role summary
dataset 14
background 3
citation-polarity summary
representative citing papers
Integrating direction-of-arrival spectra and binaural embeddings from passive audio into vision models improves relative camera pose estimation in in-the-wild videos and adds robustness to visual corruption.
C4G introduces compact timestamp-conditioned Gaussian query tokens that aggregate full temporal context to decode 3D Gaussians with timestamp-modulated positions for feed-forward 4D reconstruction from monocular video, plus a diffusion-based rendering module and extension to 4D feature fields.
Mamba-VGGT introduces a Sliding Window Mamba memory module and Zero-Init Spatial Memory Injector to enable persistent long-range geometric reasoning in VGGT for extended video sequences.
DEVIS-GRPO applies online policy gradients with an accumulative small-to-large view sampling strategy and multi-level rewards to improve trajectory-controlled extreme view video generation, reporting gains on Kubric-4D, iPhone, and DL3DV datasets.
GTA generates 3D worlds from single images via a two-stage video diffusion process that prioritizes geometry before appearance to improve structural consistency.
AdaptSplat adds a Frequency-Preserving Adapter to vision foundation models to boost high-frequency fidelity and cross-domain performance in feed-forward 3D Gaussian Splatting.
SplatWeaver uses cardinality Gaussian experts and pixel-level routing to dynamically allocate varying numbers of Gaussian primitives for generalizable novel view synthesis.
AirZoo is a new large-scale synthetic dataset for aerial 3D vision that improves state-of-the-art models on image retrieval, cross-view matching, and 3D reconstruction when used for fine-tuning.
GSCompleter completes 3DGS scenes from sparse viewpoints using a generate-then-register workflow with stereo-anchor view selection and ray-constrained registration to achieve metric-aware results and SOTA performance on benchmarks.
URoPE is a parameter-free relative position embedding for transformers that works across arbitrary geometric spaces by ray sampling and projection, yielding consistent gains on novel view synthesis, 3D detection, tracking, and depth estimation.
GlobalSplat achieves competitive novel-view synthesis on RealEstate10K and ACID using only 16K Gaussians via global scene tokens and coarse-to-fine training, with a 4MB footprint and under 78ms inference.
TokenGS uses learnable Gaussian tokens in an encoder-decoder architecture to regress 3D means directly, achieving SOTA feed-forward reconstruction on static and dynamic scenes with better robustness.
CT-1 transfers spatial reasoning from vision-language models to estimate camera trajectories, which are then used in a video diffusion model with wavelet regularization to produce controllable videos, claiming 25.7% better accuracy than prior methods.
Video diffusion models can be adapted into permutation-invariant generators for sparse novel view synthesis by treating the problem as video completion and removing temporal order cues.
Mem3R achieves better long-sequence 3D reconstruction by decoupling tracking and mapping with a hybrid memory of TTT-updated MLP and explicit tokens, reducing model size and trajectory errors.
OmniCamera disentangles video content and camera motion for multi-task generation with arbitrary camera control via the OmniCAM hybrid dataset and Dual-level Curriculum Co-Training.
CamReasoner uses structured O-T-A reasoning and RL on 56k samples to lift camera movement classification from 73.8% to 78.4% and VQA from 60.9% to 74.5% on Qwen2.5-VL-7B.
π³ is a feed-forward network with full permutation equivariance that outputs affine-invariant poses and scale-invariant local point maps without reference frames, reaching state-of-the-art on camera pose, depth, and dense reconstruction benchmarks.
DFoT enables flexible history conditioning in video diffusion, with history guidance methods that boost temporal consistency and support long rollouts.
RayDer is a unified transformer backbone for self-supervised static-scene novel view synthesis that absorbs dynamic content as a nuisance factor and shows power-law scaling with data and compute while matching supervised methods in zero-shot settings.
CameraNoise embeds camera motion into the noise space of video diffusion via Geometry-guided Reprojection Flow and noise warping to achieve faithful trajectory control while preserving the diffusion prior.
Standard VLMs achieve expert-level 3D performance on depth estimation, pose estimation, and object understanding via three simple techniques without architecture changes or regression losses.
FGQ applies diagonal Fisher information to guide learnable affine transformations in PTQ for multi-task VGGT, yielding up to 39% relative gains over baselines at 4-bit quantization.
citing papers explorer
-
SpatialBench: Is Your Spatial Foundation Model an All-Round Player?
SpatialBench evaluates 41 spatial foundation models across 6 paradigms and 5 task suites, finds they are not all-round players, and introduces the DA-Next-5M dataset plus DA-Next baseline model.
-
Audio-Visual Camera Pose Estimation with Passive Scene Sounds and In-the-Wild Video
Integrating direction-of-arrival spectra and binaural embeddings from passive audio into vision models improves relative camera pose estimation in in-the-wild videos and adds robustness to visual corruption.
-
Learning Global Motion with Compact Gaussians for Feed-Forward 4D Reconstruction
C4G introduces compact timestamp-conditioned Gaussian query tokens that aggregate full temporal context to decode 3D Gaussians with timestamp-modulated positions for feed-forward 4D reconstruction from monocular video, plus a diffusion-based rendering module and extension to 4D feature fields.
-
Mamba-VGGT: Persistent Long-Sequence Video Geometry Grounded Transformer via External Sliding Window Mamba Memory
Mamba-VGGT introduces a Sliding Window Mamba memory module and Zero-Init Spatial Memory Injector to enable persistent long-range geometric reasoning in VGGT for extended video sequences.
-
DEVIS-GRPO: Unleashing GRPO on Dynamic Extreme View Synthesis
DEVIS-GRPO applies online policy gradients with an accumulative small-to-large view sampling strategy and multi-level rewards to improve trajectory-controlled extreme view video generation, reporting gains on Kubric-4D, iPhone, and DL3DV datasets.
-
GTA: Advancing Image-to-3D World Generation via Geometry Then Appearance Video Diffusion
GTA generates 3D worlds from single images via a two-stage video diffusion process that prioritizes geometry before appearance to improve structural consistency.
-
AdaptSplat: Adapting Vision Foundation Models for Feed-Forward 3D Gaussian Splatting
AdaptSplat adds a Frequency-Preserving Adapter to vision foundation models to boost high-frequency fidelity and cross-domain performance in feed-forward 3D Gaussian Splatting.
-
SplatWeaver: Learning to Allocate Gaussian Primitives for Generalizable Novel View Synthesis
SplatWeaver uses cardinality Gaussian experts and pixel-level routing to dynamically allocate varying numbers of Gaussian primitives for generalizable novel view synthesis.
-
AirZoo: A Unified Large-Scale Dataset for Grounding Aerial Geometric 3D Vision
AirZoo is a new large-scale synthetic dataset for aerial 3D vision that improves state-of-the-art models on image retrieval, cross-view matching, and 3D reconstruction when used for fine-tuning.
-
GSCompleter: A Distillation-Free Plugin for Metric-Aware 3D Gaussian Splatting Completion in Seconds
GSCompleter completes 3DGS scenes from sparse viewpoints using a generate-then-register workflow with stereo-anchor view selection and ray-constrained registration to achieve metric-aware results and SOTA performance on benchmarks.
-
URoPE: Universal Relative Position Embedding across Geometric Spaces
URoPE is a parameter-free relative position embedding for transformers that works across arbitrary geometric spaces by ray sampling and projection, yielding consistent gains on novel view synthesis, 3D detection, tracking, and depth estimation.
-
GlobalSplat: Efficient Feed-Forward 3D Gaussian Splatting via Global Scene Tokens
GlobalSplat achieves competitive novel-view synthesis on RealEstate10K and ACID using only 16K Gaussians via global scene tokens and coarse-to-fine training, with a 4MB footprint and under 78ms inference.
-
TokenGS: Decoupling 3D Gaussian Prediction from Pixels with Learnable Tokens
TokenGS uses learnable Gaussian tokens in an encoder-decoder architecture to regress 3D means directly, achieving SOTA feed-forward reconstruction on static and dynamic scenes with better robustness.
-
CT-1: Vision-Language-Camera Models Transfer Spatial Reasoning Knowledge to Camera-Controllable Video Generation
CT-1 transfers spatial reasoning from vision-language models to estimate camera trajectories, which are then used in a video diffusion model with wavelet regularization to produce controllable videos, claiming 25.7% better accuracy than prior methods.
-
Novel View Synthesis as Video Completion
Video diffusion models can be adapted into permutation-invariant generators for sparse novel view synthesis by treating the problem as video completion and removing temporal order cues.
-
Mem3R: Streaming 3D Reconstruction with Hybrid Memory via Test-Time Training
Mem3R achieves better long-sequence 3D reconstruction by decoupling tracking and mapping with a hybrid memory of TTT-updated MLP and explicit tokens, reducing model size and trajectory errors.
-
OmniCamera: A Unified Framework for Multi-task Video Generation with Arbitrary Camera Control
OmniCamera disentangles video content and camera motion for multi-task generation with arbitrary camera control via the OmniCAM hybrid dataset and Dual-level Curriculum Co-Training.
-
CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning
CamReasoner uses structured O-T-A reasoning and RL on 56k samples to lift camera movement classification from 73.8% to 78.4% and VQA from 60.9% to 74.5% on Qwen2.5-VL-7B.
-
$\pi^3$: Permutation-Equivariant Visual Geometry Learning
π³ is a feed-forward network with full permutation equivariance that outputs affine-invariant poses and scale-invariant local point maps without reference frames, reaching state-of-the-art on camera pose, depth, and dense reconstruction benchmarks.
-
History-Guided Video Diffusion
DFoT enables flexible history conditioning in video diffusion, with history guidance methods that boost temporal consistency and support long rollouts.
-
RayDer: Scalable Self-Supervised Novel View Synthesis from Real-World Video
RayDer is a unified transformer backbone for self-supervised static-scene novel view synthesis that absorbs dynamic content as a nuisance factor and shows power-law scaling with data and compute while matching supervised methods in zero-shot settings.
-
CameraNoise: Enabling Faithful Camera Control in Video Diffusion through Geometry-Flow-Guided Noise Warping
CameraNoise embeds camera motion into the noise space of video diffusion via Geometry-guided Reprojection Flow and noise warping to achieve faithful trajectory control while preserving the diffusion prior.
-
VLM3: Vision Language Models Are Native 3D Learners
Standard VLMs achieve expert-level 3D performance on depth estimation, pose estimation, and object understanding via three simple techniques without architecture changes or regression losses.
-
Not All Tasks Quantize Equally: Fisher-Guided Quantization for Visual Geometry Transformer
FGQ applies diagonal Fisher information to guide learnable affine transformations in PTQ for multi-task VGGT, yielding up to 39% relative gains over baselines at 4-bit quantization.
-
Warp-as-History: Generalizable Camera-Controlled Video Generation from One Training Video
Warp-as-History enables zero-shot camera trajectory following in frozen video models by supplying camera-warped pseudo-history, with single-video LoRA fine-tuning improving generalization to unseen videos.
-
$h$-control: Training-Free Camera Control via Block-Conditional Gibbs Refinement
h-control augments hard-replacement guidance with block-conditional pseudo-Gibbs refinement on unobserved latent sites and adaptive 3D patch freezing to achieve superior FVD on RealEstate10K and DAVIS.
-
Divide and Conquer: Decoupled Representation Alignment for Multimodal World Models
M²-REPA decouples modality-specific features inside a diffusion model and aligns each to its matching expert foundation model via an alignment loss plus a decoupling regularizer, yielding better visual quality and long-term consistency in multi-modal video generation.
-
Long-tail Internet photo reconstruction
Finetuning 3D foundation models on simulated sparse subsets from MegaDepth-X produces robust reconstructions from extremely sparse, noisy internet photos while preserving performance on dense benchmarks.
-
Memorize When Needed: Decoupled Memory Control for Spatially Consistent Long-Horizon Video Generation
A decoupled memory branch with hybrid cues, cross-attention, and gating improves spatial consistency and data efficiency in long-horizon camera-trajectory video generation.
-
Self-Improving 4D Perception via Self-Distillation
SelfEvo enables pretrained 4D perception models to self-improve on unlabeled videos via self-distillation, delivering up to 36.5% relative gains in video depth estimation and 20.1% in camera estimation across eight benchmarks.
-
SurfelSplat: Learning Efficient and Generalizable Gaussian Surfel Representations for Sparse-View Surface Reconstruction
A feed-forward model regresses accurate Gaussian surfel geometry from sparse views using Nyquist-guided cross-view feature aggregation, achieving 100x speedup over optimization-based 3DGS surface methods on DTU benchmarks.
-
INSPATIO-WORLD: A Real-Time 4D World Simulator via Spatiotemporal Autoregressive Modeling
INSPATIO-WORLD is a real-time framework for high-fidelity 4D scene generation and navigation from monocular videos via STAR architecture with implicit caching, explicit geometric constraints, and distribution-matching distillation.
-
NavCrafter: Exploring 3D Scenes from a Single Image
NavCrafter generates controllable novel-view videos from one image via video diffusion, geometry-aware expansion, and enhanced 3D Gaussian Splatting to achieve state-of-the-art synthesis under large viewpoint changes.
-
HD-VGGT: High-Resolution Visual Geometry Transformer
HD-VGGT achieves state-of-the-art high-resolution 3D reconstruction from image collections via a dual-branch architecture that predicts coarse geometry at low resolution and refines details at high resolution while modulating unreliable features.
-
C3G: Learning Compact 3D Representations with 2K Gaussians
C3G creates compact 3D Gaussian representations with 2K points by guiding placement via learnable tokens that aggregate multi-view features through attention, yielding better efficiency and performance than dense methods.
-
Speculative Coupled Decoding for Training-Free Lossless Acceleration of Autoregressive Visual Generation
Speculative Coupled Decoding stabilizes draft sampling in Speculative Jacobi Decoding via an information-theoretic coupling step, delivering up to 4.2x image and 13.6x video speedups with no quality loss or training.
-
Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling
Geometry Forcing aligns video diffusion representations with geometric foundation model features via angular cosine and scale regression objectives to improve 3D consistency in generated videos.
-
BulletGen: Improving 4D Reconstruction with Bullet-Time Generation
BulletGen enhances 4D dynamic scene reconstruction from monocular videos by supervising Gaussian optimization with diffusion-generated frames aligned at a bullet-time step, achieving SOTA on novel-view synthesis and tracking.
-
CamCo: Camera-Controllable 3D-Consistent Image-to-Video Generation
CamCo equips image-to-video generators with Plücker-coordinate camera inputs and epipolar attention to improve 3D consistency and camera controllability.
-
CameraCtrl: Enabling Camera Control for Text-to-Video Generation
CameraCtrl enables accurate camera pose control in video diffusion models through a trained plug-and-play module and dataset choices emphasizing diverse camera trajectories with matching appearance.
-
RealLiFe: Real-Time Light Field Reconstruction via Hierarchical Sparse Gradient Descent
RealLiFe optimizes multi-plane images with HSGD to deliver real-time light field reconstruction from sparse views, claiming 100x speedup over offline methods and 2 dB PSNR gain over online ones.
-
LangFlash: Feed-forward 3D Language Gaussian Splatting from Sparse Unposed Images
LangFlash introduces a feed-forward model for 3D language Gaussian splatting from sparse unposed images, claiming superior novel view synthesis and semantic consistency via enriched training data and sparse semantic encoding.
-
IVGT: Implicit Visual Geometry Transformer for Neural Scene Representation
IVGT implicitly models continuous neural scene representations from pose-free multi-view images to enable coherent surface extraction, novel view synthesis, and related 3D tasks via SDF and color prediction.
-
SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer
SANA-WM is a 2.6B-parameter efficient world model that synthesizes minute-scale 720p videos with 6-DoF camera control, trained on 213K public clips in 15 days on 64 H100s and runnable on single GPUs at 36x higher throughput than prior open baselines.
-
RoSplat: Robust Feed-Forward Pixel-wise Gaussian Splatting for Varying Input Views and High-Resolution Rendering
RoSplat adds alpha normalization for brightness consistency across varying input views and a 3D sampling regularizer to mitigate hole artifacts in high-resolution feed-forward Gaussian splatting.
-
ST-Gen4D: Embedding 4D Spatiotemporal Cognition into World Model for 4D Generation
ST-Gen4D uses a world model that fuses global appearance and local dynamic graphs into a 4D cognition representation to guide consistent 4D Gaussian generation.
-
Video Generation with Predictive Latents
PV-VAE improves video latent spaces for generation by unifying reconstruction with future-frame prediction, reporting 52% faster convergence and 34.42 FVD gain over Wan2.2 VAE on UCF101.
-
Learning 3D Representations for Spatial Intelligence from Unposed Multi-View Images
UniSplat learns consistent 3D geometry, appearance, and semantics from unposed images using dual masking, progressive Gaussian splatting, and recalibration to align predictions across tasks.
-
InSpatio-WorldFM: An Open-Source Real-Time Generative Frame Model
InSpatio-WorldFM is a frame-independent generative model that uses explicit 3D anchors and spatial memory to deliver real-time multi-view consistent spatial intelligence via a three-stage training pipeline from pretrained diffusion models.
-
Hestia: Voxel-Face-Aware Hierarchical Next-Best-View Acquisition for Efficient 3D Reconstruction
Hestia improves generalizable next-best-view planning for 3D reconstruction via hierarchical action search, diverse data, close-greedy strategy, and face-aware voxel design, yielding higher coverage and lower Chamfer distance than prior RL-based methods.