WarpHammer densifies scene warps with 3D object priors from generative models and fuses pose-unknown auxiliary views via multi-view geometry to enable stable extreme novel view synthesis.
hub Mixed citations
Lvsm: A large view synthesis model with minimal 3d inductive bias
Mixed citation behavior. Most common role is background (67%).
23
Pith papers citing it
Background
67% of classified citations
abstract
We propose the Large View Synthesis Model (LVSM), a novel transformer-based approach for scalable and generalizable novel view synthesis from sparse-view inputs. We introduce two architectures: (1) an encoder-decoder LVSM, which encodes input image tokens into a fixed number of 1D latent tokens, functioning as a fully learned scene representation, and decodes novel-view images from them; and (2) a decoder-only LVSM, which directly maps input images to novel-view outputs, completely eliminating intermediate scene representations. Both models bypass the 3D inductive biases used in previous methods -- from 3D representations (e.g., NeRF, 3DGS) to network designs (e.g., epipolar projections, plane sweeps) -- addressing novel view synthesis with a fully data-driven approach. While the encoder-decoder model offers faster inference due to its independent latent representation, the decoder-only LVSM achieves superior quality, scalability, and zero-shot generalization, outperforming previous state-of-the-art methods by 1.5 to 3.5 dB PSNR. Comprehensive evaluations across multiple datasets demonstrate that both LVSM variants achieve state-of-the-art novel view synthesis quality. Notably, our models surpass all previous methods even with reduced computational resources (1-2 GPUs). Please see our website for more details: https://haian-jin.github.io/projects/LVSM/ .
hub tools
citation-role summary
background 4
baseline 2
citation-polarity summary
representative citing papers
FLAT maps compressed video diffusion latents to explicit triangle splats via ray-centered rotation parameterization and a product window function, reporting better geometric accuracy than 3D Gaussian baselines under identical training.
A kinematic-to-visual lifting paradigm combined with hierarchically routed control generates action-conditioned surgical videos with better faithfulness, fidelity, and efficiency.
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.
A 3D-grounded autoencoder and diffusion transformer allow direct generation of 3D scenes in an implicit latent space using a fixed 1K-token representation for arbitrary views and resolutions.
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.
ART is a category-agnostic transformer that maps sparse multi-state RGB images to per-part 3D geometry, texture, and articulation parameters via learnable part slots.
FastForward represents scenes as collections of 3D-anchored image features and performs camera pose estimation via feed-forward correspondence prediction, achieving competitive accuracy with minimal mapping time.
Anchoring Gaussian centers to predicted raymaps and jointly optimizing RGB, raymap, and camera losses with a dual-frequency curriculum suppresses pose drift and improves pose-free 3D reconstruction on long sequences.
DPPE decouples rotation and translation in camera positional encodings for multi-view transformers to resolve late-stage training stagnation and improve generalization in novel view synthesis.
Diffusion-based per-view harmonization for lighting-consistent object transfer between 3DGS scenes, using heterogeneous training data and final 3D consolidation.
GHOST improves generalization in robot manipulation via hierarchical factorization into 3D sub-goal prediction from RGB-D views and a goal-conditioned low-level controller, enabling human video integration without action retargeting.
A transformer-based neural renderer that transfers arbitrary PBR lighting to single images via shared intrinsic conditioning extracted from both multi-illumination photos and path-traced coarse 3D renders.
FreeScale generates scalable high-quality training data for generalizable novel view synthesis by certainty-aware sampling from imperfect scene reconstructions, delivering 2.7 dB PSNR gains on out-of-distribution tests.
A feed-forward transformer predicts SMPL-X vertex-aligned 3D Gaussians in a canonical T-pose, enabling real-time animation by linear blend skinning without per-frame network inference.
Scaling transformer context with sparse attention and 3D-aware block routing improves feed-forward 3D reconstruction and inverse rendering, closing much of the quality gap with dense-view optimization.
Large-chunk online updates during inference let test-time training scale state capacity to 40% of model size and handle contexts up to 1M tokens without custom kernels.
RR2D reconstructs a virtual analog-domain covariance from partially observed hybrid array data via Toeplitz-constrained matrix completion, enabling practical hybrid SMI beamforming that outperforms direct hybrid SMI by 3-4 dB in SINR.
Long-LRM++ achieves real-time 14 FPS high-fidelity 360-degree scene reconstruction from 32-64 views by using semi-explicit Gaussians plus a light decoder, matching LaCT quality on DL3DV and improving depth prediction.
ViPE estimates camera intrinsics, motion, and dense near-metric depth from uncalibrated videos, outperforming baselines on TUM and KITTI while releasing annotations for 96M frames across real and generated videos.
OpenWorldLib defines world models as perception-centered systems with interaction and long-term memory, and provides a modular inference codebase unifying interactive video, 3D, reasoning, and VLA tasks.
The book presents principles from optimization and information theory to explain deep network architectures and enable new interpretable models.
citing papers explorer
-
WarpHammer: Densifying Scene Warps with 3D Object Priors for Extreme View Synthesis
WarpHammer densifies scene warps with 3D object priors from generative models and fuses pose-unknown auxiliary views via multi-view geometry to enable stable extreme novel view synthesis.
-
FLAT: Feedforward Latent Triangle Splatting for Geometrically Accurate Scene Generation
FLAT maps compressed video diffusion latents to explicit triangle splats via ray-centered rotation parameterization and a product window function, reporting better geometric accuracy than 3D Gaussian baselines under identical training.
-
From Articulated Kinematics to Routed Visual Control for Action-Conditioned Surgical Video Generation
A kinematic-to-visual lifting paradigm combined with hierarchically routed control generates action-conditioned surgical videos with better faithfulness, fidelity, and efficiency.
-
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.
-
Any 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale
A 3D-grounded autoencoder and diffusion transformer allow direct generation of 3D scenes in an implicit latent space using a fixed 1K-token representation for arbitrary views and resolutions.
-
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.
-
ART: Articulated Reconstruction Transformer
ART is a category-agnostic transformer that maps sparse multi-state RGB images to per-part 3D geometry, texture, and articulation parameters via learnable part slots.
-
A Scene is Worth a Thousand Features: Feed-Forward Camera Localization from a Collection of Image Features
FastForward represents scenes as collections of 3D-anchored image features and performs camera pose estimation via feed-forward correspondence prediction, achieving competitive accuracy with minimal mapping time.
-
NoDrift3R: Raymap-Guided Coupling for Drift-Robust Unposed Feed-Forward 3D Reconstruction
Anchoring Gaussian centers to predicted raymaps and jointly optimizing RGB, raymap, and camera losses with a dual-frequency curriculum suppresses pose drift and improves pose-free 3D reconstruction on long sequences.
-
DPPE: Rethinking Camera-Based Positional Encoding for Scaling Multi-View Transformers
DPPE decouples rotation and translation in camera positional encodings for multi-view transformers to resolve late-stage training stagnation and improve generalization in novel view synthesis.
-
Lighting-Consistent Object Transfer Across Radiance Fields
Diffusion-based per-view harmonization for lighting-consistent object transfer between 3DGS scenes, using heterogeneous training data and final 3D consolidation.
-
GHOST: Hierarchical Sub-Goal Policies for Generalizing Robot Manipulation
GHOST improves generalization in robot manipulation via hierarchical factorization into 3D sub-goal prediction from RGB-D views and a goal-conditioned low-level controller, enabling human video integration without action retargeting.
-
PIXLRelight: Controllable Relighting via Intrinsic Conditioning
A transformer-based neural renderer that transfers arbitrary PBR lighting to single images via shared intrinsic conditioning extracted from both multi-illumination photos and path-traced coarse 3D renders.
-
FreeScale: Scaling 3D Scenes via Certainty-Aware Free-View Generation
FreeScale generates scalable high-quality training data for generalizable novel view synthesis by certainty-aware sampling from imperfect scene reconstructions, delivering 2.7 dB PSNR gains on out-of-distribution tests.
-
Real-Time Human Reconstruction and Animation using Feed-Forward Gaussian Splatting
A feed-forward transformer predicts SMPL-X vertex-aligned 3D Gaussians in a canonical T-pose, enabling real-time animation by linear blend skinning without per-frame network inference.
-
LSRM: High-Fidelity Object-Centric Reconstruction via Scaled Context Windows
Scaling transformer context with sparse attention and 3D-aware block routing improves feed-forward 3D reconstruction and inverse rendering, closing much of the quality gap with dense-view optimization.
-
Test-Time Training Done Right
Large-chunk online updates during inference let test-time training scale state capacity to 40% of model size and handle contexts up to 1M tokens without custom kernels.
-
URoPE: Universal Relative Position Embedding across Geometric Spaces
RR2D reconstructs a virtual analog-domain covariance from partially observed hybrid array data via Toeplitz-constrained matrix completion, enabling practical hybrid SMI beamforming that outperforms direct hybrid SMI by 3-4 dB in SINR.
-
Long-LRM++: Preserving Fine Details in Feed-Forward Wide-Coverage Reconstruction
Long-LRM++ achieves real-time 14 FPS high-fidelity 360-degree scene reconstruction from 32-64 views by using semi-explicit Gaussians plus a light decoder, matching LaCT quality on DL3DV and improving depth prediction.
-
ViPE: Video Pose Engine for 3D Geometric Perception
ViPE estimates camera intrinsics, motion, and dense near-metric depth from uncalibrated videos, outperforming baselines on TUM and KITTI while releasing annotations for 96M frames across real and generated videos.
-
OpenWorldLib: A Unified Codebase and Definition of Advanced World Models
OpenWorldLib defines world models as perception-centered systems with interaction and long-term memory, and provides a modular inference codebase unifying interactive video, 3D, reasoning, and VLA tasks.
-
Principles and Practice of Deep Representation Learning: or a Mathematical Theory of Memory
The book presents principles from optimization and information theory to explain deep network architectures and enable new interpretable models.