CrossFit-GS: Out-of-Fold Pose Estimation for COLMAP-Free 3D Gaussian Splatting

Zijian Wu; Jinliang Wang; Mingfeng Jiang; Zidian Lin; Ying Song; Ziqian Lu

cs.CVOct 20, 2026

CrossFit-GS: Out-of-Fold Pose Estimation for COLMAP-Free 3D Gaussian Splatting

Authors: Zijian Wu, Jinliang Wang, Mingfeng Jiang, Zidian Lin, Ying Song, Ziqian Lu

Organizations: Zhejiang Sci-Tech University · Tsinghua University

Abstract

Pose-free 3D Gaussian Splatting (3DGS) aims to bypass cumbersome Structure-from-Motion (SfM) by directly reconstructing scenes from unposed image sequences. In this work, we present CrossFit-GS, a deferred-commitment progressive framework that mitigates cumulative trajectory drift and breaks the self-reinforcing loop of distorted-map error propagation. Specifically, we partition the input sequence into interleaved subsets and maintain two complementary submaps. Each candidate pose is cross-refined against a frozen opposite submap that has never been optimized using the target frame, thereby providing an independent geometric reference for pose estimation. Checkerboard validation and spatial coverage safeguards further reject unreliable updates, and only verified poses are committed to their corresponding owner submap. Once the trajectory has been refined, the intermediate submaps are discarded and the final global representation is trained from scratch using the fixed refined poses. Extensive evaluations across multiple benchmarks demonstrate that CrossFit-GS substantially reduces cumulative trajectory drift and consistently improves both camera pose estimation and novel-view synthesis over existing methods.

Figures & tables

Figure 2: Overview of CrossFit-GS. The input sequence is partitioned into two interleaved folds, each associated with a separate Gaussian map. For a frame in Fold B, its pose is refined against the frozen Map A, with visual-support checks and an accept-or-keep validation rule governing the update. The frame subsequently contributes only to Map B. The roles of the two folds alternate, ensuring that a frame does not train the map used to assess its pose. Finally, a unified Gaussian representation is trained from scratch using all training images and the fixed estimated poses.

Scene	CF-3DGS [ 9 ]			HT-3DGS [ 15 ]			TrackGS [ 33 ]			PCR-GS [ 44 ]			Ours
Scene	PSNR	SSIM	LPIPS	PSNR	SSIM	LPIPS	PSNR	SSIM	LPIPS	PSNR	SSIM	LPIPS	PSNR	SSIM	LPIPS
Ballroom	24.917	0.828	0.123	25.620	0.850	0.110	25.709	0.875	0.105	25.164	0.832	0.117	30.272	0.931	0.060
Barn	19.264	0.567	0.382	19.327	0.560	0.354	21.623	0.625	0.280	19.003	0.534	0.458	22.863	0.657	0.262
Church	26.825	0.869	0.142	31.046	0.942	0.073	25.960	0.831	0.206	27.308	0.885	0.133	29.246	0.922	0.092
Family	25.033	0.845	0.164	27.801	0.889	0.124	28.663	0.899	0.119	27.147	0.896	0.119	29.079	0.918	0.104
Francis	23.898	0.737	0.270	24.313	0.717	0.257	24.365	0.738	0.272	24.774	0.727	0.258	25.359	0.742	0.266

Table 1: Quantitative novel view synthesis results on Tanks and Temples. Evaluated using PSNR

\uparrow

, SSIM

\uparrow

, and LPIPS

\downarrow

. Best results are bolded.

Figure 3: Qualitative comparison of novel-view synthesis on Tanks and Temples. CrossFit-GS preserves fine textures and sharp structural boundaries while reducing rendering artifacts.

Sequence	CF-3DGS [ 9 ]			HT-3DGS [ 15 ]			PCR-GS [ 44 ]			TrackGS [ 33 ]			Ours
Sequence	PSNR	SSIM	LPIPS	PSNR	SSIM	LPIPS	PSNR	SSIM	LPIPS	PSNR	SSIM	LPIPS	PSNR	SSIM	LPIPS
apple_23361	31.758	0.899	0.134	30.477	0.859	0.163	34.305	0.935	0.106	34.059	0.935	0.103	35.034	0.939	0.101
hydrant_108483	21.852	0.724	0.314	22.337	0.759	0.279	22.129	0.722	0.317	20.822	0.658	0.320	23.432	0.779	0.263
plant_50907	25.931	0.744	0.310	27.463	0.795	0.283	25.554	0.733	0.316	23.178	0.643	0.389	30.543	0.859	0.235
teddybear_38541	32.048	0.862	0.334	33.957	0.900	0.279	32.059	0.862	0.333	33.509	0.865	0.288	34.732	0.903	0.290
bench_110109	32.509	0.894	0.144	30.302	0.844	0.195	31.397	0.856	0.164	27.415	0.713	0.277	34.595	0.933	0.109

Table 2: Quantitative NVS results on selected CO3D-V2 sequences. Evaluated using PSNR

\uparrow

, SSIM

\uparrow

, and LPIPS

\downarrow

. Best results are bolded.

Figure 4: Qualitative comparison of novel-view synthesis on CO3D-V2. CrossFit-GS delivers highly accurate geometry and sharp visual details under rapid camera rotations, effectively suppressing blur and floating artifacts.

Figure 5: Visualization of estimated camera trajectories. The top row shows the Ignatius sequence from Tanks and Temples, and the bottom row shows the apple_23361 sequence from CO3D-V2. Red solid lines denote estimated trajectories, while blue dashed lines indicate the ground truth.

Method	Tanks and Temples			CO3D-V2
Method	$\text{RPE}_{t}$	$\text{RPE}_{r}$	ATE	$\text{RPE}_{t}$	$\text{RPE}_{r}$	ATE
CF-3DGS [ 9 ]	1.327	0.695	0.024	0.515	0.336	0.018
HT-3DGS [ 15 ]	1.229	0.667	0.022	0.460	0.324	0.019
TrackGS [ 33 ]	0.600	0.348	0.014	0.992	0.782	0.041
PCR-GS [ 44 ]	1.440	0.596	0.041	0.453	0.331	0.014
Ours	0.571	0.347	0.010	0.344	0.323	0.009

Table 3: Quantitative camera pose estimation results on Tanks and Temples (stride 5) and selected CO3D-V2 sequences. Evaluated using

\text{RPE}_{t}\downarrow(\times 10^{-2})

\text{RPE}_{r}\downarrow(^{\circ})

, and

\text{ATE}\downarrow

. Best results are bolded.

Variant	Pose			NVS
Variant	$\text{RPE}_{t}$	$\text{RPE}_{r}$	ATE	PSNR	SSIM	LPIPS
Local tracking only	1.327	0.695	0.024	21.931	0.707	0.252
Naive cross-fit	0.636	0.381	0.017	25.486	0.801	0.174
w/o Coverage Gate	0.643	0.370	0.016	25.835	0.813	0.170
w/o Retention Guard	0.578	0.367	0.014	25.909	0.812	0.167
w/o Checkerboard	0.612	0.362	0.015	25.733	0.810	0.167

Table 4: Ablation study on Tanks and Temples. Evaluated across pose metrics (

\text{RPE}_{t}\downarrow\times 10^{-2}

\text{RPE}_{r}\downarrow^{\circ}

\text{ATE}\downarrow

) and NVS metrics (

\text{PSNR}\uparrow

\text{SSIM}\uparrow

\text{LPIPS}\downarrow

). Best results are bolded.

Method	Total Time (min) $\downarrow$
HT-3DGS [ 15 ]	91.0
TrackGS [ 33 ]	88.0
CF-3DGS [ 9 ]	64.0
Ours	67.5

Table 5: Efficiency comparison on the first 50 frames of selected CO3D-V2 sequences. Total runtime is measured in minutes (m).

Variant	Target-Used	Map Frozen	Validation Loss $\downarrow$	ATE $\downarrow$	PSNR $\uparrow$
TCTM	Yes	No	0.000849	0.028	23.702
TCFM	Yes	Yes	0.000850	0.027	23.789
TETM	No	No	0.005092	0.029	23.305
TEFM (Ours)	No	Yes	0.032814	0.010	26.246

Table 6: Diagnostic ablation of the self-explanation pathway on Tanks and Temples. Target-Used : whether the target image is used to train the reference map before pose refinement. Map Frozen : whether the reference map parameters are frozen during pose refinement.

3D Gaussian Splatting

3D Gaussian Splatting represents scenes as optimizable 3D Gaussian primitives for fast, photorealistic rendering from new viewpoints.

See all 1,013 papers

SG-Splat: Stable Geometry Learning under Pose Uncertainty for Pose-Free 3D Gaussian Splatting
Oct 26, 2026Xiaohan Lu, Jiaqi Zhang, Weijia Jiang +13D Gaussian Splatting Feed-Forward 3D Reconstruction
State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University, Beijing 100871, China · Migu Interactive Entertainment Co., Ltd, Nanjing 210000, China
DeltaSplat: Iterative Gaussian Refinement for Pose-Free Feed-Forward 3D Gaussian Splatting
Oct 7, 2026Chanung Park, Seunghyeon Song, Joo Chan Lee +23D Gaussian Splatting 3D Reconstruction
Department of Electrical and Computer Engineering, Sungkyunkwan University · Department of Artificial Intelligence, Sungkyunkwan University · Department of Artificial Intelligence, Yonsei University
Reliability-Regulated Trajectory Optimization for Progressive COLMAP-Free 3D Gaussian Splatting
Sep 25, 2026Zijian Wu, Jinliang Wang, Zidian Lin +53D Gaussian Splatting Camera Pose Estimation
School of Computer Science and Technology, Zhejiang Sci-Tech University · School of Biomedical Engineering, Tsinghua University · CardioCloud Medical Technology (Beijing) Co., Ltd. +3

Oct 26, 2026J/K move · Enter open · S save

SG-Splat: Stable Geometry Learning under Pose Uncertainty for Pose-Free 3D Gaussian Splatting

Xiaohan Lu, Jiaqi Zhang, Weijia Jiang, Siwei Ma

State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University, Beijing 100871, China · Migu Interactive Entertainment Co., Ltd, Nanjing 210000, China

3D Gaussian Splatting Feed-Forward 3D Reconstruction Novel View Synthesis Camera Pose Estimation

Open paper AskPDF arXiv

3D Gaussian Splatting (3DGS) has become a promising representation for real-time novel view synthesis and 3D scene reconstruction. For practical deployment, recent studies have explored pose-free feed-forward 3DGS, which reconstructs Gaussian representations directly from unposed input images. However, these methods remain sensitive to residual pose errors, which perturb canonical-space Gaussian transformation, introducing noisy supervision for geometry and attribute learning. To address this problem, we present SG-Splat, a pose-free feed-forward 3DGS framework that improves robustness to pose uncertainty and stabilizes the learning of Gaussian primitives. SG-Splat decouples geometry estimation and Gaussian attribute prediction into two dedicated branches, reducing error coupling between pose-sensitive geometry and Gaussian attributes. We further introduce a coarse-to-fine refinement pipeline, consisting of Flow-Guided Pose Refinement (FGPR) and Residual-Guided Geometry Correction (RGC), to improve geometric alignment before Gaussian attribute reconstruction. Specifically, FGPR refines camera poses with flow-based correlation features and differentiable bundle adjustment. RGC then corrects initial point maps using depth residuals derived from FGPR, producing more reliable Gaussian centers. To stabilize training, we introduce a dual-path strategy that mixes ground-truth and predicted poses for canonical transformation. It provides clean supervision for Gaussian attributes while improving geometry robustness to residual pose errors. Extensive experiments on DL3DV and RealEstate10K show that SG-Splat achieves state-of-the-art performance in both novel view synthesis and pose estimation.

Abstract

3D Gaussian Splatting

Explore similar work

SG-Splat: Stable Geometry Learning under Pose Uncertainty for Pose-Free 3D Gaussian Splatting

DeltaSplat: Iterative Gaussian Refinement for Pose-Free Feed-Forward 3D Gaussian Splatting

Reliability-Regulated Trajectory Optimization for Progressive COLMAP-Free 3D Gaussian Splatting