Accurate 3D reconstruction of vehicles is vital for applications such as vehicle inspection, predictive maintenance, and urban planning. Existing methods like Neural Radiance Fields and Gaussian Splatting have shown impressive results but remain limited by their reliance on dense input views, which hinders real-world applicability. This paper addresses the challenge of reconstructing vehicles from sparse-view inputs, leveraging depth maps and a robust pose estimation architecture to synthesize novel views and augment training data. Specifically, we enhance Gaussian Splatting by integrating a selective photometric loss, applied only to high-confidence pixels, and replacing standard Structure-from-Motion pipelines with the DUSt3R architecture to improve camera pose estimation. Furthermore, we present a novel dataset featuring both synthetic and real-world public transportation vehicles, enabling extensive evaluation of our approach. Experimental results demonstrate state-of-the-art performance across multiple benchmarks, showcasing the method's ability to achieve high-quality reconstructions even under constrained input conditions.
We introduce $\pi^3$, a feed-forward neural network that offers a novel approach to visual geometry reconstruction, breaking the reliance on a conventional fixed reference view. Previous methods often anchor their reconstructions to a designated viewpoint, an inductive bias that can lead to instability and failures if the reference is suboptimal. In contrast, $\pi^3$ employs a fully permutation-equivariant architecture to predict affine-invariant camera poses and scale-invariant local point maps without any reference frames. This design makes our model inherently robust to input ordering and highly scalable. These advantages enable our simple and bias-free approach to achieve state-of-the-art performance on a wide range of tasks, including camera pose estimation, monocular/video depth estimation, and dense point map reconstruction. Code and models are publicly available.
We propose SGLoc, a novel localization system that directly regresses camera poses from 3D Gaussian Splatting (3DGS) representation by leveraging semantic information. Our method utilizes the semantic relationship between 2D image and 3D scene representation to estimate the 6DoF pose without prior pose information. In this system, we introduce a multi-level pose regression strategy that progressively estimates and refines the pose of query image from the global 3DGS map, without requiring initial pose priors. Moreover, we introduce a semantic-based global retrieval algorithm that establishes correspondences between 2D (image) and 3D (3DGS map). By matching the extracted scene semantic descriptors of 2D query image and 3DGS semantic representation, we align the image with the local region of the global 3DGS map, thereby obtaining a coarse pose estimation. Subsequently, we refine the coarse pose by iteratively optimizing the difference between the query image and the rendered image from 3DGS. Our SGLoc demonstrates superior performance over baselines on 12scenes and 7scenes datasets, showing excellent capabilities in global localization without initial pose prior. Code will be available at https://github.com/IRMVLab/SGLoc.
We present SpatialTrackerV2, a feed-forward 3D point tracking method for monocular videos. Going beyond modular pipelines built on off-the-shelf components for 3D tracking, our approach unifies the intrinsic connections between point tracking, monocular depth, and camera pose estimation into a high-performing and feedforward 3D point tracker. It decomposes world-space 3D motion into scene geometry, camera ego-motion, and pixel-wise object motion, with a fully differentiable and end-to-end architecture, allowing scalable training across a wide range of datasets, including synthetic sequences, posed RGB-D videos, and unlabeled in-the-wild footage. By learning geometry and motion jointly from such heterogeneous data, SpatialTrackerV2 outperforms existing 3D tracking methods by 30%, and matches the accuracy of leading dynamic 3D reconstruction approaches while running 50$\times$ faster.
Camera pose estimation is a fundamental computer vision task that is essential for applications like visual localization and multi-view stereo reconstruction. In the object-centric scenarios with sparse inputs, the accuracy of pose estimation can be significantly influenced by background textures that occupy major portions of the images across different viewpoints. In light of this, we introduce the Kaleidoscopic Background Attack (KBA), which uses identical segments to form discs with multi-fold radial symmetry. These discs maintain high similarity across different viewpoints, enabling effective attacks on pose estimation models even with natural texture segments. Additionally, a projected orientation consistency loss is proposed to optimize the kaleidoscopic segments, leading to significant enhancement in the attack effectiveness. Experimental results show that optimized adversarial kaleidoscopic backgrounds can effectively attack various camera pose estimation models.
Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have emerged as powerful tools for 3D reconstruction and SLAM tasks. However, their performance depends heavily on accurate camera pose priors. Existing approaches attempt to address this issue by introducing external constraints but fall short of achieving satisfactory accuracy, particularly when camera trajectories are complex. In this paper, we propose a novel method, RA-NeRF, capable of predicting highly accurate camera poses even with complex camera trajectories. Following the incremental pipeline, RA-NeRF reconstructs the scene using NeRF with photometric consistency and incorporates flow-driven pose regulation to enhance robustness during initialization and localization. Additionally, RA-NeRF employs an implicit pose filter to capture the camera movement pattern and eliminate the noise for pose estimation. To validate our method, we conduct extensive experiments on the Tanks\&Temple dataset for standard evaluation, as well as the NeRFBuster dataset, which presents challenging camera pose trajectories. On both datasets, RA-NeRF achieves state-of-the-art results in both camera pose estimation and visual quality, demonstrating its effectiveness and robustness in scene reconstruction under complex pose trajectories.
We introduce Princeton365, a large-scale diverse dataset of 365 videos with accurate camera pose. Our dataset bridges the gap between accuracy and data diversity in current SLAM benchmarks by introducing a novel ground truth collection framework that leverages calibration boards and a 360-camera. We collect indoor, outdoor, and object scanning videos with synchronized monocular and stereo RGB video outputs as well as IMU. We further propose a new scene scale-aware evaluation metric for SLAM based on the the optical flow induced by the camera pose estimation error. In contrast to the current metrics, our new metric allows for comparison between the performance of SLAM methods across scenes as opposed to existing metrics such as Average Trajectory Error (ATE), allowing researchers to analyze the failure modes of their methods. We also propose a challenging Novel View Synthesis benchmark that covers cases not covered by current NVS benchmarks, such as fully non-Lambertian scenes with 360-degree camera trajectories. Please visit https://princeton365.cs.princeton.edu for the dataset, code, videos, and submission.
Real-world applications like video gaming and virtual reality often demand the ability to model 3D scenes that users can explore along custom camera trajectories. While significant progress has been made in generating 3D objects from text or images, creating long-range, 3D-consistent, explorable 3D scenes remains a complex and challenging problem. In this work, we present Voyager, a novel video diffusion framework that generates world-consistent 3D point-cloud sequences from a single image with user-defined camera path. Unlike existing approaches, Voyager achieves end-to-end scene generation and reconstruction with inherent consistency across frames, eliminating the need for 3D reconstruction pipelines (e.g., structure-from-motion or multi-view stereo). Our method integrates three key components: 1) World-Consistent Video Diffusion: A unified architecture that jointly generates aligned RGB and depth video sequences, conditioned on existing world observation to ensure global coherence 2) Long-Range World Exploration: An efficient world cache with point culling and an auto-regressive inference with smooth video sampling for iterative scene extension with context-aware consistency, and 3) Scalable Data Engine: A video reconstruction pipeline that automates camera pose estimation and metric depth prediction for arbitrary videos, enabling large-scale, diverse training data curation without manual 3D annotations. Collectively, these designs result in a clear improvement over existing methods in visual quality and geometric accuracy, with versatile applications.
We propose the first 4D tracking and mapping method that jointly performs camera localization and non-rigid surface reconstruction via differentiable rendering. Our approach captures 4D scenes from an online stream of color images with depth measurements or predictions by jointly optimizing scene geometry, appearance, dynamics, and camera ego-motion. Although natural environments exhibit complex non-rigid motions, 4D-SLAM remains relatively underexplored due to its inherent challenges; even with 2.5D signals, the problem is ill-posed because of the high dimensionality of the optimization space. To overcome these challenges, we first introduce a SLAM method based on Gaussian surface primitives that leverages depth signals more effectively than 3D Gaussians, thereby achieving accurate surface reconstruction. To further model non-rigid deformations, we employ a warp-field represented by a multi-layer perceptron (MLP) and introduce a novel camera pose estimation technique along with surface regularization terms that facilitate spatio-temporal reconstruction. In addition to these algorithmic challenges, a significant hurdle in 4D SLAM research is the lack of reliable ground truth and evaluation protocols, primarily due to the difficulty of 4D capture using commodity sensors. To address this, we present a novel open synthetic dataset of everyday objects with diverse motions, leveraging large-scale object models and animation modeling. In summary, we open up the modern 4D-SLAM research by introducing a novel method and evaluation protocols grounded in modern vision and rendering techniques.
Currently almost all state-of-the-art novel view synthesis and reconstruction models rely on calibrated cameras or additional geometric priors for training. These prerequisites significantly limit their applicability to massive uncalibrated data. To alleviate this requirement and unlock the potential for self-supervised training on large-scale uncalibrated videos, we propose a novel two-stage strategy to train a view synthesis model from only raw video frames or multi-view images, without providing camera parameters or other priors. In the first stage, we learn to reconstruct the scene implicitly in a latent space without relying on any explicit 3D representation. Specifically, we predict per-frame latent camera and scene context features, and employ a view synthesis model as a proxy for explicit rendering. This pretraining stage substantially reduces the optimization complexity and encourages the network to learn the underlying 3D consistency in a self-supervised manner. The learned latent camera and implicit scene representation have a large gap compared with the real 3D world. To reduce this gap, we introduce the second stage training by explicitly predicting 3D Gaussian primitives. We additionally apply explicit Gaussian Splatting rendering loss and depth projection loss to align the learned latent representations with physically grounded 3D geometry. In this way, Stage 1 provides a strong initialization and Stage 2 enforces 3D consistency - the two stages are complementary and mutually beneficial. Extensive experiments demonstrate the effectiveness of our approach, achieving high-quality novel view synthesis and accurate camera pose estimation, compared to methods that employ supervision with calibration, pose, or depth information. The code is available at https://github.com/Dwawayu/Pensieve.
Accurate and robust 3D scene reconstruction from casual, in-the-wild videos can significantly simplify robot deployment to new environments. However, reliable camera pose estimation and scene reconstruction from such unconstrained videos remains an open challenge. Existing visual-only SLAM methods perform well on benchmark datasets but struggle with real-world footage which often exhibits uncontrolled motion including rapid rotations and pure forward movements, textureless regions, and dynamic objects. We analyze the limitations of current methods and introduce a robust pipeline designed to improve 3D reconstruction from casual videos. We build upon recent deep visual odometry methods but increase robustness in several ways. Camera intrinsics are automatically recovered from the first few frames using structure-from-motion. Dynamic objects and less-constrained areas are masked with a predictive model. Additionally, we leverage monocular depth estimates to regularize bundle adjustment, mitigating errors in low-parallax situations. Finally, we integrate place recognition and loop closure to reduce long-term drift and refine both intrinsics and pose estimates through global bundle adjustment. We demonstrate large-scale contiguous 3D models from several online videos in various environments. In contrast, baseline methods typically produce locally inconsistent results at several points, producing separate segments or distorted maps. In lieu of ground-truth pose data, we evaluate map consistency, execution time and visual accuracy of re-rendered NeRF models. Our proposed system establishes a new baseline for visual reconstruction from casual uncontrolled videos found online, demonstrating more consistent reconstructions over longer sequences of in-the-wild videos than previously achieved.
Dynamic videos with small baseline motions are ubiquitous in daily life, especially on social media. However, these videos present a challenge to existing pose estimation frameworks due to ambiguous features, drift accumulation, and insufficient triangulation constraints. Gaussian splatting, which maintains an explicit representation for scenes, provides a reliable novel view rasterization when the viewpoint change is small. Inspired by this, we propose SmallGS, a camera pose estimation framework that is specifically designed for small-baseline videos. SmallGS optimizes sequential camera poses using Gaussian splatting, which reconstructs the scene from the first frame in each video segment to provide a stable reference for the rest. The temporal consistency of Gaussian splatting within limited viewpoint differences reduced the requirement of sufficient depth variations in traditional camera pose estimation. We further incorporate pretrained robust visual features, e.g. DINOv2, into Gaussian splatting, where high-dimensional feature map rendering enhances the robustness of camera pose estimation. By freezing the Gaussian splatting and optimizing camera viewpoints based on rasterized features, SmallGS effectively learns camera poses without requiring explicit feature correspondences or strong parallax motion. We verify the effectiveness of SmallGS in small-baseline videos in TUM-Dynamics sequences, which achieves impressive accuracy in camera pose estimation compared to MonST3R and DORID-SLAM for small-baseline videos in dynamic scenes. Our project page is at: https://yuxinyao620.github.io/SmallGS
Recent advancements in autonomous driving (AD) systems have highlighted the potential of world models in achieving robust and generalizable performance across both ordinary and challenging driving conditions. However, a key challenge remains: precise and flexible camera pose control, which is crucial for accurate viewpoint transformation and realistic simulation of scene dynamics. In this paper, we introduce PosePilot, a lightweight yet powerful framework that significantly enhances camera pose controllability in generative world models. Drawing inspiration from self-supervised depth estimation, PosePilot leverages structure-from-motion principles to establish a tight coupling between camera pose and video generation. Specifically, we incorporate self-supervised depth and pose readouts, allowing the model to infer depth and relative camera motion directly from video sequences. These outputs drive pose-aware frame warping, guided by a photometric warping loss that enforces geometric consistency across synthesized frames. To further refine camera pose estimation, we introduce a reverse warping step and a pose regression loss, improving viewpoint precision and adaptability. Extensive experiments on autonomous driving and general-domain video datasets demonstrate that PosePilot significantly enhances structural understanding and motion reasoning in both diffusion-based and auto-regressive world models. By steering camera pose with self-supervised depth, PosePilot sets a new benchmark for pose controllability, enabling physically consistent, reliable viewpoint synthesis in generative world models.
Sparse-view camera pose estimation, which aims to estimate the 6-Degree-of-Freedom (6-DoF) poses from a limited number of images captured from different viewpoints, is a fundamental yet challenging problem in remote sensing applications. Existing methods often overlook the translation information between each pair of viewpoints, leading to suboptimal performance in sparse-view scenarios. To address this limitation, we introduce T-Graph, a lightweight, plug-and-play module to enhance camera pose estimation in sparse-view settings. T-graph takes paired image features as input and maps them through a Multilayer Perceptron (MLP). It then constructs a fully connected translation graph, where nodes represent cameras and edges encode their translation relationships. It can be seamlessly integrated into existing models as an additional branch in parallel with the original prediction, maintaining efficiency and ease of use. Furthermore, we introduce two pairwise translation representations, relative-t and pair-t, formulated under different local coordinate systems. While relative-t captures intuitive spatial relationships, pair-t offers a rotation-disentangled alternative. The two representations contribute to enhanced adaptability across diverse application scenarios, further improving our module's robustness. Extensive experiments on two state-of-the-art methods (RelPose++ and Forge) using public datasets (C03D and IMC PhotoTourism) validate both the effectiveness and generalizability of T-Graph. The results demonstrate consistent improvements across various metrics, notably camera center accuracy, which improves by 1% to 6% from 2 to 8 viewpoints.
Traditional SLAM systems, which rely on bundle adjustment, struggle with highly dynamic scenes commonly found in casual videos. Such videos entangle the motion of dynamic elements, undermining the assumption of static environments required by traditional systems. Existing techniques either filter out dynamic elements or model their motion independently. However, the former often results in incomplete reconstructions, whereas the latter can lead to inconsistent motion estimates. Taking a novel approach, this work leverages a 3D point tracker to separate the camera-induced motion from the observed motion of dynamic objects. By considering only the camera-induced component, bundle adjustment can operate reliably on all scene elements as a result. We further ensure depth consistency across video frames with lightweight post-processing based on scale maps. Our framework combines the core of traditional SLAM -- bundle adjustment -- with a robust learning-based 3D tracker front-end. Integrating motion decomposition, bundle adjustment and depth refinement, our unified framework, BA-Track, accurately tracks the camera motion and produces temporally coherent and scale-consistent dense reconstructions, accommodating both static and dynamic elements. Our experiments on challenging datasets reveal significant improvements in camera pose estimation and 3D reconstruction accuracy.
Multi-view understanding, the ability to reconcile visual information across diverse viewpoints for effective navigation, manipulation, and 3D scene comprehension, is a fundamental challenge in Multi-Modal Large Language Models (MLLMs) to be used as embodied agents. While recent MLLMs have shown impressive advances in high-level reasoning and planning, they frequently fall short when confronted with multi-view geometric consistency and cross-view correspondence. To comprehensively evaluate the challenges of MLLMs in multi-view scene reasoning, we propose All-Angles Bench, a benchmark of over 2,100 human carefully annotated multi-view question-answer pairs across 90 diverse real-world scenes. Our six tasks (counting, attribute identification, relative distance, relative direction, object manipulation, and camera pose estimation) specifically test model's geometric correspondence and the capacity to align information consistently across views. Our extensive experiments, benchmark on 27 representative MLLMs including Gemini-2.0-Flash, Claude-3.7-Sonnet, and GPT-4o against human evaluators reveals a substantial performance gap, indicating that current MLLMs remain far from human-level proficiency. Through in-depth analysis, we show that MLLMs are particularly underperforming under two aspects: (1) cross-view correspondence for partially occluded views and (2) establishing the coarse camera poses. These findings highlight the necessity of domain-specific refinements or modules that embed stronger multi-view awareness. We believe that our All-Angles Bench offers valuable insights and contribute to bridging the gap between MLLMs and human-level multi-view understanding. The project and benchmark are publicly available at https://danielchyeh.github.io/All-Angles-Bench/.
Recent advances in data-driven geometric multi-view 3D reconstruction foundation models (e.g., DUSt3R) have shown remarkable performance across various 3D vision tasks, facilitated by the release of large-scale, high-quality 3D datasets. However, as we observed, constrained by their matching-based principles, the reconstruction quality of existing models suffers significant degradation in challenging regions with limited matching cues, particularly in weakly textured areas and low-light conditions. To mitigate these limitations, we propose to harness the inherent robustness of monocular geometry estimation to compensate for the inherent shortcomings of matching-based methods. Specifically, we introduce a monocular-guided refinement module that integrates monocular geometric priors into multi-view reconstruction frameworks. This integration substantially enhances the robustness of multi-view reconstruction systems, leading to high-quality feed-forward reconstructions. Comprehensive experiments across multiple benchmarks demonstrate that our method achieves substantial improvements in both mutli-view camera pose estimation and point cloud accuracy.
Multi-view 3D reconstruction has remained an essential yet challenging problem in the field of computer vision. While DUSt3R and its successors have achieved breakthroughs in 3D reconstruction from unposed images, these methods exhibit significant limitations when scaling to multi-view scenarios, including high computational cost and cumulative error induced by global alignment. To address these challenges, we propose Regist3R, a novel stereo foundation model tailored for efficient and scalable incremental reconstruction. Regist3R leverages an incremental reconstruction paradigm, enabling large-scale 3D reconstructions from unordered and many-view image collections. We evaluate Regist3R on public datasets for camera pose estimation and 3D reconstruction. Our experiments demonstrate that Regist3R achieves comparable performance with optimization-based methods while significantly improving computational efficiency, and outperforms existing multi-view reconstruction models. Furthermore, to assess its performance in real-world applications, we introduce a challenging oblique aerial dataset which has long spatial spans and hundreds of views. The results highlight the effectiveness of Regist3R. We also demonstrate the first attempt to reconstruct large-scale scenes encompassing over thousands of views through pointmap-based foundation models, showcasing its potential for practical applications in large-scale 3D reconstruction tasks, including urban modeling, aerial mapping, and beyond.
3D Gaussian Splatting (3DGS) is a powerful reconstruction technique, but it needs to be initialized from accurate camera poses and high-fidelity point clouds. Typically, the initialization is taken from Structure-from-Motion (SfM) algorithms; however, SfM is time-consuming and restricts the application of 3DGS in real-world scenarios and large-scale scene reconstruction. We introduce a constrained optimization method for simultaneous camera pose estimation and 3D reconstruction that does not require SfM support. Core to our approach is decomposing a camera pose into a sequence of camera-to-(device-)center and (device-)center-to-world optimizations. To facilitate, we propose two optimization constraints conditioned to the sensitivity of each parameter group and restricts each parameter's search space. In addition, as we learn the scene geometry directly from the noisy point clouds, we propose geometric constraints to improve the reconstruction quality. Experiments demonstrate that the proposed method significantly outperforms the existing (multi-modal) 3DGS baseline and methods supplemented by COLMAP on both our collected dataset and two public benchmarks.
Reconstructing 3D scenes from monocular surgical videos can enhance surgeon's perception and therefore plays a vital role in various computer-assisted surgery tasks. However, achieving scale-consistent reconstruction remains an open challenge due to inherent issues in endoscopic videos, such as dynamic deformations and textureless surfaces. Despite recent advances, current methods either rely on calibration or instrument priors to estimate scale, or employ SfM-like multi-stage pipelines, leading to error accumulation and requiring offline optimization. In this paper, we present Endo3R, a unified 3D foundation model for online scale-consistent reconstruction from monocular surgical video, without any priors or extra optimization. Our model unifies the tasks by predicting globally aligned pointmaps, scale-consistent video depths, and camera parameters without any offline optimization. The core contribution of our method is expanding the capability of the recent pairwise reconstruction model to long-term incremental dynamic reconstruction by an uncertainty-aware dual memory mechanism. The mechanism maintains history tokens of both short-term dynamics and long-term spatial consistency. Notably, to tackle the highly dynamic nature of surgical scenes, we measure the uncertainty of tokens via Sampson distance and filter out tokens with high uncertainty. Regarding the scarcity of endoscopic datasets with ground-truth depth and camera poses, we further devise a self-supervised mechanism with a novel dynamics-aware flow loss. Abundant experiments on SCARED and Hamlyn datasets demonstrate our superior performance in zero-shot surgical video depth prediction and camera pose estimation with online efficiency. Project page: https://wrld.github.io/Endo3R/.