3DV 2025 Papers

The rapid growth of AR/VR/MR applications and cloud-based visual localization has heightened concerns over user privacy. This privacy concern has been further escalated by the ability of deep neural networks to recover detailed images of a scene from a sparse set of 3D or 2D points and their descriptors - the so-called inversion attacks. Research on privacy-preserving localization has therefore focused on preventing such attacks through geometry obfuscation techniques like lifting points to higher dimensions or swapping coordinates. In this paper, we reveal a common vulnerability in these methods that allows approximate point recovery using known neighborhoods. We further show that these neighborhoods can be computed by learning to identify descriptors that co-occur in neighborhoods. Extensive experiments demonstrate that all existing geometric obfuscation schemes remain susceptible to such recovery, challenging their claims of being privacy-preserving.

Interactions between human and objects are influenced not only by the object's pose and shape, but also by physical attributes such as object mass and surface friction. They introduce important motion nuances that are essential for diversity and realism. Despite advancements in recent human-object interaction methods, this aspect has been overlooked. Generating nuanced human motion presents two challenges. First, it is non-trivial to learn from multi-modal human and object information derived from both the physical and non-physical attributes. Second, there exists no dataset capturing nuanced human interactions with objects of varying physical properties, hampering model development. This work addresses the gap by introducing the FORCE model, an approach for synthesizing diverse, nuanced human-object interactions by modeling physical attributes. Our key insight is that human motion is dictated by the interrelation between the force exerted by the human and the perceived resistance. Guided by a novel intuitive physics encoding, the model captures the interplay between human force and resistance. Experiments also demonstrate incorporating human force facilitates learning multi-class motion. Accompanying our model, we contribute the a dataset, which features diverse, different-styled motion through interactions with varying resistances. Our code, dataset, and models will be released to foster future research.

Visual localization involves estimating the 6-degree-of-freedom (6-DoF) camera pose within a known scene. A critical step in this process is identifying pixel-to-point correspondences between 2D query images and 3D models. Most advanced approaches currently rely on extensive visual descriptors to establish these correspondences, facing challenges in storage, privacy issues and model maintenance. Direct 2D-3D keypoint matching without visual descriptors is becoming popular as it can overcome those challenges. However, existing descriptor-free methods suffer from low accuracy or heavy computation. Addressing this gap, this paper introduces the Angle-Annular Graph Neural Network (A2-GNN), a simple approach that efficiently learns robust geometric structural representations with annular feature extraction. Specifically, this approach clusters neighbors and embeds each group’s distance information and angle as supplementary information to capture local structures. Evaluation on matching and visual localization datasets demonstrates that our approach achieves state-of-the-art accuracy with low computational overhead among visual description-free methods.

Owing to the distinctive attributes inherent to mmWave radar, the utilization of millimeter-wave (mmWave) point cloud data in contexts involving human-related scenarios is poised to yield significant promise. However, the generation of dense and temporally consistent 3D human point clouds from sequential mmWave signals is a challenging task, yet a critical endeavor with far-reaching implications. In this work, we present a groundbreaking approach to address the challenge of generating dense and temporally consistent 3D human point clouds from sequential mmWave signals. We redefine the problem as a 3D point cloud denoising task, leveraging reverse diffusion processes to transform sparse mmWave data into detailed human representations. Our proposed method, mmDiffusion, effectively exploits diffusion models and temporal context within mmWave sequences to learn the denoising process, resulting in denser and temporally coherent human point clouds. We also introduce a novel evaluation metric tailored to measure temporal consistency. Experimental results demonstrate that mmDiffusion outperforms existing methods. Code and dataset will be public upon acceptance.

Effectively modeling the interaction between human hands and objects is challenging due to the complex physical constraints and the equirement for high generation efficiency in applications. Prior approaches often employ computationally intensive two-stage approaches, which first generate an intermediate representation, such as contact maps, followed by an iterative optimization procedure that updates hand meshes to capture the hand-object relation. However, due to the high computation complexity during the optimization stage, such strategies often suffer from low efficiency in inference. To address this limitation, this work introduces a novel diffusion-model-based approach that generates the grasping pose in a one-stage manner. This allows us to significantly improve generation speed and the diversity of generated hand poses. In particular, we develop a Latent Diffusion Model with an Adaptation Module for object-conditioned hand pose generation and a contact-aware loss to enforce the physical constraints between hands and objects. Extensive experiments demonstrate that our method achieves faster inference, higher diversity, and superior pose quality than state-of-the-art approaches. Code is available at https://github.com/wuxiaofei01/FastGrasp .

Camera rotation estimation from a single image is a challenging task, often requiring depth data and/or camera intrinsics, which are generally not available for in-the-wild videos. Although external sensors such as inertial measurement units (IMUs) can help, they often suffer from drift and are not applicable in non-inertial reference frames. We present U-ARE-ME, an algorithm that estimates camera rotation along with uncertainty from uncalibrated RGB images. Using a Manhattan World assumption, our method leverages the per-pixel geometric priors encoded in single-image surface normal predictions and performs optimisation over the SO(3) manifold. Given a sequence of images, we can use the per-frame rotation estimates and their uncertainty to perform multi-frame optimisation, achieving robustness and temporal consistency. Our experiments demonstrate that U-ARE-ME performs comparably to RGB-D methods and is more robust than feature-based vanishing point and SLAM methods.

Synthesizing accurate geometry and photo-realistic appearance of small scenes is an active area of research with compelling use cases in gaming, virtual reality, robotic-manipulation, autonomous driving, convenient product capture, and consumer-level photography. When applying scene geometry and appearance estimation techniques to robotics, we found that the narrow cone of possible viewpoints due to the limited range of robot motion and scene clutter caused current estimation techniques to produce poor quality estimates or even fail. On the other hand, in robotic applications, dense metric depth can often be measured directly using stereo and illumination can be controlled. Depth can provide a good initial estimate of the object geometry to improve reconstruction, while multi-illumination images can facilitate relighting. In this work we demonstrate a method to incorporate dense metric depth into the training of neural 3D representations and address an artifact observed while jointly refining geometry and appearance by disambiguating between texture and geometry edges. We also discuss a multi-flash stereo camera system developed in-house to capture the necessary data for our pipeline and show results on relighting and view synthesis with a few training views.

We present iFusion, a novel 3D object reconstruction framework that requires only two views with unknown camera poses. While single-view reconstruction yields visually appealing results, it can deviate significantly from the actual object, especially on unseen sides. Additional views improve reconstruction fidelity but necessitate known camera poses. However, assuming the availability of pose may be unrealistic, and existing pose estimators fail in sparse-view scenarios. To address this, we harness a pre-trained novel view synthesis diffusion model, which embeds implicit knowledge about the geometry and appearance of diverse objects. Our strategy unfolds in three steps: (1) We invert the diffusion model for camera pose estimation instead of synthesizing novel views. (2) The diffusion model is fine-tuned using provided views and estimated poses, turned into a novel view synthesizer tailored for the target object. (3) Leveraging registered views and the fine-tuned diffusion model, we reconstruct the 3D object. Experiments demonstrate strong performance in both pose estimation and novel view synthesis. Moreover, iFusion seamlessly integrates with various reconstruction methods and enhances them.

This paper presents a novel framework for compactly representing a 3D indoor scene using a set of polycuboids through a deep learning-based fitting method. Indoor scenes mainly consist of man-made objects, such as furniture, which often exhibit rectilinear geometry. This property allows indoor scenes to be represented using combinations of polycuboids, providing a compact representation that benefits downstream applications like furniture rearrangement. Our framework takes a noisy point cloud as input and first detects six types of cuboid faces using a transformer network. Then, a graph neural network is used to validate the spatial relationships of the detected faces to form potential polycuboids. Finally, each polycuboid instance is reconstructed by forming a set of boxes based on the aggregated face labels. To train our networks, we introduce a synthetic dataset encompassing a diverse range of cuboid and polycuboid shapes that reflect the characteristics of indoor scenes. Our framework generalizes well to real-world indoor scene datasets, including Replica, ScanNet, and scenes captured with an iPhone. The versatility of our method is demonstrated through practical applications, such as virtual room tours and scene editing.

Learning-based methods address its inherent scale ambiguity by leveraging increasingly large labeled and unlabeled datasets, to produce geometric priors capable of generating accurate predictions across domains. As a result, state of the art approaches show impressive performance in zero-shot relative and metric depth estimation. Recently, diffusion models have exhibited remarkable scalability and generalizable properties in their learned representations. However, because these models repurpose tools originally designed for image generation, they can only operate on dense ground-truth, which is not available for most depth labels, especially in real-world settings. In this paper we present GRIN, an efficient diffusion model designed to ingest sparse unstructured training data. We use image features with 3D geometric positional encodings to condition the diffusion process both globally and locally, generating depth predictions at a pixel-level. With comprehensive experiments across eight indoor and outdoor datasets, we show that GRIN establishes a new state of the art in zero-shot metric monocular depth estimation even when trained from scratch.

We present a method for reconstructing a clear Neural Radiance Field (NeRF) even with fast camera motions. To address blur artifacts, we leverage both (blurry) RGB images and event camera data captured in a binocular configuration. Importantly, when reconstructing our clear NeRF, we consider the camera modeling imperfections that arise from the simple pinhole camera model as learned embeddings for each camera measurement, and further learn a mapper that connects event camera measurements with RGB data. As no previous dataset exists for our binocular setting, we introduce an event camera dataset with captures from a 3D-printed stereo configuration between RGB and event cameras. Empirically, we evaluate on our introduced dataset and EVIMOv2 and show that our method leads to improved reconstructions. We are committed to making our code and dataset public.

We introduce SAMPro3D for zero-shot instance segmentation of 3D scenes. Given the 3D point cloud and multiple posed RGB-D frames of 3D scenes, our approach segments 3D instances by applying the pretrained Segment Anything Model (SAM) to 2D frames. Our key idea involves locating SAM prompts in 3D to align their projected pixel prompts across frames, ensuring the view consistency of SAM-predicted masks. Moreover, we suggest selecting prompts from the initial set guided by the information of SAM-predicted masks across all views, which enhances the overall performance. We further propose to consolidate different prompts if they are segmenting different surface parts of the same 3D instance, bringing a more comprehensive segmentation. Notably, our method does **not** require any additional training. Extensive experiments on diverse benchmarks show that our method achieves comparable or better performance compared to previous zero-shot or fully supervised approaches, and in many cases surpasses human annotations. Furthermore, since our fine-grained predictions often lack annotations in available datasets, we present ScanNet200-Fine50 test data which provides more fine-grained annotations on 50 scenes from ScanNet200 dataset.

This paper introduces a robust translation synchronization approach which takes relative directions between pairs of images as inputs and outputs absolute image translations. Our approach is based on a generalized eigenvalue problem, where the formulation contains edge weights in relative directions and vertex weights in absolute image translations. We present a rigorous stability analysis to determine how to set these weights optimally. Specifically, optimal vertex weights are always $1$, while optimal edge weights depend on magnitudes of relative transformations and variances of relative directions. These results lead to an iterative translation synchronization formulation, which progressively removes outliers in the inputs by adaptively adjusting the edge weights. We present exact and robust recovery conditions for our approach under a standard noise model. Experimental results justify our theoretical results and show that our approach outperforms state-of-the-art baseline approaches on both synthetic and real datasets.

We introduce a new approach for generating realistic 3D models with UV maps through a representation termed "Object Images." This approach encapsulates surface geometry, appearance, and patch structures within a 64x64 pixel image, effectively converting complex 3D shapes into a more manageable 2D format. By doing so, we address the challenges of both geometric and semantic irregularity inherent in polygonal meshes. This method allows us to use image generation models, such as Diffusion Transformers, directly for 3D shape generation. Evaluated on the ABO dataset, our generated shapes with patch structures achieve point cloud FID comparable to recent 3D generative models, while naturally supporting PBR material generation.

The latest regularized Neural Radiance Field (NeRF) approaches produce poor geometry and view extrapolation for large scale sparse view scenes, such as ETH3D. Density-based approaches tend to be under-constrained, while surface-based approaches tend to miss details. In this paper, we take a density-based approach, sampling patches instead of individual rays to better incorporate monocular depth and normal estimates and patch-based photometric consistency constraints between training views and sampled virtual views. Loosely constraining densities based on estimated depth aligned to sparse points further improves geometric accuracy. While maintaining similar view synthesis quality, our approach significantly improves geometric accuracy on the ETH3D benchmark, e.g. increasing the F1@2cm score by 4x-8x compared to other regularized density-based approaches, with much lower training and inference time than other approaches.

Point clouds are a set of data points in space to represent the 3D geometry of objects. A fundamental step in the processing is to identify a subset of points to represent the shape. While traditional sampling methods often ignore to incorporate geometrical information, recent developments in learning-based sampling models have achieved significant levels of performance. With the integration of geometrical priors, the ability to learn and preserve the underlying structure can be enhanced when sampling. To shed light into the shape, a qualitative skeleton serves as an effective descriptor to guide sampling for both local and global geometries. In this paper, we introduce MorphoSkel3D as a new technique based on morphology to facilitate an efficient skeletonization of shapes. With its low computational cost, MorphoSkel3D is a unique, rule-based algorithm to benchmark its quality and performance on two large datasets, ModelNet and ShapeNet, under different sampling ratios. The results show that training with MorphoSkel3D leads to an informed and more accurate sampling in the practical application of object classification and point cloud retrieval.

We present a spatio-temporal perspective on category-agnostic 3D lifting of 2D keypoints over a temporal sequence. Our approach differs from existing state-of-the-art methods that are either: (i) object-agnostic, but can only operate on individual frames, or (ii) can model space-time dependencies, but are only designed to work with a single object category. Our approach is grounded in two core principles. First, general information about similar objects can be leveraged to achieve better performance when there is little object-specific training data. Second, a temporally-proximate context window is advantageous for achieving consistency throughout a sequence. These two principles allow us to outperform current state-of-the-art methods on per-frame and per-sequence metrics for a variety of animal categories. Lastly, we release a new synthetic dataset containing 3D skeletons and motion sequences for a variety of animal categories.

This paper introduces FlowMap, an end-to-end differentiable method that solves for precise camera poses, camera intrinsics, and per-frame dense depth of a video sequence. Our method performs per-video gradient-descent minimization of a simple least-squares objective that compares the optical flow induced by depth, intrinsics, and poses against correspondences obtained via off-the-shelf optical flow and point tracking. Alongside the use of point tracks to encourage long-term geometric consistency, we introduce a differentiable re-parameterization of depth, intrinsics, and pose that is amenable to first-order optimization. We empirically show that camera parameters and dense depth recovered by our method enable photo-realistic novel view synthesis on 360° trajectories using Gaussian Splatting. Our method not only far outperforms prior gradient-descent based bundle adjustment methods, but surprisingly performs on par with COLMAP, the state-of-the-art SfM method, on the downstream task of 360° novel view synthesis - even though our method is purely gradient-descent based, fully differentiable, and presents a complete departure from conventional SfM. Our result opens the door to the self-supervised training of neural networks that perform camera parameter estimation, 3D reconstruction, and novel view synthesis.

3D Gaussian Splatting (3D-GS) has recently attracted great attention with real-time and photo-realistic renderings. This technique typically takes perspective images as input and optimizes a set of 3D elliptical Gaussians by splatting them onto the image planes, resulting in 2D Gaussians. However, applying 3D-GS to panoramic inputs presents challenges in effectively modeling the projection onto the spherical surface of ${360^\circ}$ images using 2D Gaussians. In practical applications, input panoramas are often sparse, leading to unreliable initialization of 3D Gaussians and subsequent degradation of 3D-GS quality. In addition, due to the under-constrained geometry of texture-less planes (e.g., walls and floors), 3D-GS struggles to model these flat regions with elliptical Gaussians, resulting in significant floaters in novel views. To address these issues, we propose 360-GS, a novel layout-guided $360^{\circ}$ Gaussian splatting for a limited set of panoramic inputs. Instead of splatting 3D Gaussians directly onto the spherical surface, 360-GS projects them onto the tangent plane of the unit sphere and then maps them to the spherical projections. This adaptation enables the representation of the projection using Gaussians. We guide the optimization of 3D Gaussians by exploiting layout priors within panoramas, which are simple to obtain and contain strong structural information about the indoor scene. Our experimental results demonstrate that 360-GS allows panoramic rendering and outperforms state-of-the-art methods with fewer artifacts in novel view synthesis, thus providing immersive roaming in indoor scenarios.