REVIEW 39 cited by
No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
We introduce NoPoSplat, a feed-forward model capable of reconstructing 3D scenes parameterized by 3D Gaussians from \textit{unposed} sparse multi-view images. Our model, trained exclusively with photometric loss, achieves real-time 3D Gaussian reconstruction during inference. To eliminate the need for accurate pose input during reconstruction, we anchor one input view's local camera coordinates as the canonical space and train the network to predict Gaussian primitives for all views within this space. This approach obviates the need to transform Gaussian primitives from local coordinates into a global coordinate system, thus avoiding errors associated with per-frame Gaussians and pose estimation. To resolve scale ambiguity, we design and compare various intrinsic embedding methods, ultimately opting to convert camera intrinsics into a token embedding and concatenate it with image tokens as input to the model, enabling accurate scene scale prediction. We utilize the reconstructed 3D Gaussians for novel view synthesis and pose estimation tasks and propose a two-stage coarse-to-fine pipeline for accurate pose estimation. Experimental results demonstrate that our pose-free approach can achieve superior novel view synthesis quality compared to pose-required methods, particularly in scenarios with limited input image overlap. For pose estimation, our method, trained without ground truth depth or explicit matching loss, significantly outperforms the state-of-the-art methods with substantial improvements. This work makes significant advances in pose-free generalizable 3D reconstruction and demonstrates its applicability to real-world scenarios. Code and trained models are available at https://noposplat.github.io/.
Forward citations
Cited by 39 Pith papers
-
Compact Feed-Forward 3D Gaussians via Saliency-Guided Primitive Merging
A saliency-guided superpixel merging pipeline compresses feed-forward 3D Gaussian splatting outputs to about 5% of their primitives while largely preserving PSNR and SSIM.
-
ResemBrick: Brick Reconstruction from Photographs with Perceptual Fidelity and Buildability
A two-stage pipeline that learns where to spend a fixed voxel budget for best appearance and then assembles the resulting grid into bricks with zero floating or unstable components, beating prior systems on both fidel...
-
TinySplat: Feedforward Approach for Generating Compact 3D Scene Representation
TinySplat compresses feedforward 3D Gaussian scenes by 105-199x on two-view benchmarks (about 50x on DL3DV) while keeping rendered quality close to the uncompressed model.
-
GeoWorldAD: Geometry World Action Model for Autonomous Driving
Grounding an autonomous-driving action model in ego-aligned multi-scale 3D geometry and latent future-geometry tokens improves NAVSIM closed-loop PDMS/EPDMS over prior geometry- and world-model-based planners.
-
Instant NuRec: Feed-Forward 3D Gaussian Reconstruction for Driving Scene Simulation
A feed-forward model reconstructs a layered, simulation-ready 3D Gaussian world from multi-view driving video in ~1.5 s, with quality approaching per-scene optimized reconstruction.
-
AsySplat: Efficient Asymmetric 3D Gaussian Splatting for Long-Sequence Scene Modeling
An asymmetric geometry-appearance architecture for generalizable 3DGS reallocates computation so smaller models match optimization-based NVS quality at ~800× speedup on 32-view 960P inputs while improving zero-shot results.
-
MAC-Splat: Multi-Attribute Consistency for High-Fidelity Sparse-View Reconstruction
Semantically enriched MASt3R correspondences plus a multi-attribute 3D consistency loss raise sparse-view ScanNet++ PSNR by >4.5 dB over Splatt3R and preserve quality under wide baselines.
-
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.
-
WildSplat: Feedforward Gaussian Splatting from Unposed In-the-Wild Images
WildSplat decouples geometry from appearance in a single feedforward pass to produce appearance-conditioned 3D Gaussian reconstructions from unposed in-the-wild images.
-
AdaptiveSplat:Texture Aware Controllable 3D Gaussian Allocation for Feed-Forward Reconstruction
Texture-aware SuperCluster pruning plus an adaptive Gaussian head lets feed-forward 3DGS models hit a user budget β while outperforming post-hoc pruners on RE10K, ACID, DL3DV and DTU.
-
OF$^3$GS: On-the-Fly Feed-Forward 3D Gaussian Splatting from Unposed Images
FreeStreamGS achieves online NVS from unposed streaming inputs competitive with offline 3DGS methods via decoupled intrinsic recovery and dynamic point refinement.
-
AirSplat: Alignment and Rating for Robust Feed-Forward 3D Gaussian Splatting
AirSplat fine-tunes a 3D vision foundation model with a self-corrected pose loop and teacher-rated opacity pruning to achieve state-of-the-art pose-free novel view synthesis.
-
ILV: Iterative Latent Volumes for Fast and Accurate Sparse-View CT Reconstruction
ILV recovers fine anatomical detail in sparse-view CBCT by iteratively updating an explicit 3D latent volume with multi-view X-ray features and a learned prior, outperforming prior feed-forward and optimization method...
-
AnyStyle: Single-Pass Multimodal Stylization for 3D Gaussian Splatting
A single-pass 3D Gaussian splatting pipeline that stylizes unposed scenes from either a text prompt or a reference image via a lightweight zero-initialized style-injection branch.
-
MatchAttention: Embedding Explicit Matching Constraints into Attention for Efficient Stereo Matching
MatchAttention replaces global cross-attention with a linearly-scaling, learnable-offset windowed attention whose relative position field is the predicted disparity/flow, yielding state-of-the-art stereo/flow accuracy...
-
VolSplat: Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction
VolSplat predicts 3D Gaussians from a shared voxel grid instead of from image pixels, reporting large gains in sparse-view novel view synthesis on RealEstate10K, ScanNet, and ACID.
-
iLRM: An Iterative Large 3D Reconstruction Model
iLRM reconstructs 3D Gaussian scenes from multiple photos through iterative refinement of viewpoint tokens, achieving higher quality and speed than prior feed-forward models.
-
RegGS: Unposed Sparse Views Gaussian Splatting with 3DGS Registration
RegGS aligns locally generated 3D Gaussian maps using a Sinkhorn-approximated mixture Wasserstein distance, improving pose estimation and novel view synthesis from sparse unposed views.
-
Puzzles: Unbounded Video-Depth Augmentation for Scalable End-to-End 3D Reconstruction
Puzzles synthesizes posed video-depth clips from single images and keyframes, letting 3D reconstruction models match full-data accuracy using only 10% of the data.
-
UniForward: Unified 3D Scene and Semantic Field Reconstruction via Feed-Forward Gaussian Splatting from Only Sparse-View Images
UniForward is a feed-forward model that reconstructs 3D Gaussian scenes with embedded semantic features from just two unposed images, enabling real-time novel-view rendering and open-vocabulary segmentation.
-
E3D-Bench: A Benchmark for End-to-End 3D Geometric Foundation Models
E3D-Bench compares 16 3D geometric foundation models on depth, reconstruction, pose, and view-synthesis tasks with a unified evaluation toolkit.
-
X-GRM: Large Gaussian Reconstruction Model for Sparse-view X-rays to Computed Tomography
A large transformer with fixed-voxel Gaussian splatting reconstructs CT volumes from 6-10 X-ray projections in under a second, substantially beating prior sparse-view methods in simulation.
-
Dust to Tower: Coarse-to-Fine Photo-Realistic Scene Reconstruction from Sparse Uncalibrated Images
A coarse-to-fine pipeline jointly optimizes 3D Gaussian Splatting and camera poses from sparse, uncalibrated images, using warped and inpainted pseudo-views for supervision.
-
PanSplat: 4K Panorama Synthesis with Feed-Forward Gaussian Splatting
A feed-forward Gaussian splatting system that synthesizes novel 4K panoramic views from two wide-baseline inputs, using Fibonacci-lattice Gaussians and memory-efficient training.
-
LiftImage3D: Lifting Any Single Image to 3D Gaussians with Video Generation Priors
LiftImage3D generates small-motion video clips from one image, registers them with MASt3R, and fits a distortion-aware 3D Gaussian field whose canonical scene renders new views.
-
SLAM3R: Real-Time Dense Scene Reconstruction from Monocular RGB Videos
A neural system that reconstructs dense 3D scenes from monocular RGB video at 20+ FPS by regressing local pointmaps and incrementally registering them into one global model without explicit pose optimization.
-
Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localization
Reloc3r trains a symmetric, scale-free relative pose regression transformer on 8M image pairs and uses motion averaging for absolute poses, outperforming prior regression methods on six localization benchmarks.
-
SLGaussian: Fast Language Gaussian Splatting in Sparse Views
SLGaussian builds a 3D semantic field from two photos in a single forward pass, stores CLIP features in a memory bank for fast open-vocabulary queries, and reports higher IoU than LangSplat and LERF on the LERF and 3D...
-
SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting
SelfSplat jointly predicts depth, camera poses and 3D Gaussians from unposed image triplets, and outperforms prior pose-free baselines on RealEstate10K, ACID and DL3DV.
-
ZeroGS: Training 3D Gaussian Splatting from Unposed Images
A pipeline that trains 3D Gaussian Splatting from hundreds of unposed, unordered images by finetuning a pretrained pointmap foundation model and incrementally registering images.
-
RoadVGGT: Road-Structure-Aware Feed-Forward Road Surface Reconstruction
A feed-forward Gaussian head on OmniVGGT plus road-plane grid fusion and structure-aware grouping reconstructs compact road surfaces that beat RoGS and AnySplat on Waymo and zero-shot nuScenes.
-
InstanceSplat: Instance-Aware Feed-Forward 3D Gaussian Splatting for Scene Understanding
InstanceSplat predicts 3D Gaussians with per-object identity and language-aligned semantics from unposed multi-view images in a single forward pass, enabling rendering, instance segmentation, and open-vocabulary queries.
-
WorldMirror: Universal 3D World Reconstruction with Any-Prior Prompting
A unified feed-forward network that ingests optional geometric priors and jointly predicts point maps, depth, camera poses, normals, and 3D Gaussians, reporting state-of-the-art results on multiple 3D benchmarks.
-
DeblurSplat: SfM-free 3D Gaussian Splatting with Event Camera for Robust Deblurring
A pose-free deblurring 3D Gaussian Splatting pipeline using DUSt3R point clouds, confidence-balanced sampling, and event-decoded latent image supervision.
-
ProSplat: Improved Feed-Forward 3D Gaussian Splatting for Wide-Baseline Sparse Views
ProSplat combines a feed-forward 3D Gaussian Splatting generator with a one-step diffusion improvement model and epipolar-aware attention, reporting about 1 dB PSNR gain over SOTA on wide-baseline sparse-view novel vi...
-
Latent-Reframe: Enabling Camera Control for Video Diffusion Model without Training
Latent-Reframe steers a pre-trained video diffusion model along a target camera trajectory by reframing halfway-denoised latents with time-aware 3D point clouds and then inpainting the resulting gaps, all without fine-tuning.
-
Align3R: Aligned Monocular Depth Estimation for Dynamic Videos
Align3R injects monocular depth estimates into a fine-tuned DUSt3R model to recover temporally consistent video depth and camera poses for dynamic monocular videos.
-
GSGTrack: Gaussian Splatting-Guided Object Pose Tracking from RGB Videos
GSGTrack jointly optimizes Gaussian Splatting geometry and object pose to track unknown objects in RGB video, reporting large accuracy gains over SLAM baselines.
-
SmileSplat: Generalizable Gaussian Splats for Unconstrained Sparse Images
SmileSplat predicts Gaussian surfels from sparse unposed image pairs and jointly optimizes scene geometry and camera intrinsics and extrinsics, reporting state-of-the-art novel view synthesis on Re10K, ACID, Replica, ...
Discussion (0). Continue with ORCID to comment.