3DV 2025 Papers

We present DreamBeast, a novel method based on score distillation sampling (SDS) for generating fantastical 3D animal assets composed of distinct parts. Existing SDS methods often struggle with this generation task due to a limited understanding of part-level semantics in text-to-image diffusion models. While recent diffusion models, such as Stable Diffusion 3, demonstrate a better part-level understanding, they are prohibitively slow and exhibit other common problems associated with single-view diffusion models. DreamBeast overcomes this limitation through a novel part-aware knowledge transfer mechanism. For each generated asset, we efficiently extract part-level knowledge from the Stable Diffusion 3 model into a 3D part-affinity implicit representation. This enables us to instantly generate part-affinity maps from arbitrary camera views, which we then use to modulate the guidance of a multi-view diffusion model during SDS to generate 3D assets of fantastical animals. DreamBeast significantly enhances the quality of generated 3D creatures with user-specified part compositions while reducing computational overhead, as demonstrated by extensive quantitative and qualitative evaluations.

3D part segmentation is still an open problem in the field of 3D vision and AR/VR. Due to limited 3D labeled data, traditional supervised segmentation methods fall short in generalizing to unseen shapes and categories. Recently, the advancement in vision-language models' zero-shot abilities has brought a surge in open-world 3D segmentation methods. While these methods show promising results for 3D scenes or objects, they do not generalize well to 3D humans. In this paper, we present the first open-vocabulary segmentation method capable of handling 3D human. Our framework can segment the human category into desired fine-grained parts based on the textual prompt. We design a simple segmentation pipeline, leveraging SAM to generate multi-view proposals in 2D and proposing a novel HumanCLIP model to create unified embeddings for visual and textual inputs. Compared with existing pre-trained CLIP models, the HumanCLIP model yields more accurate embeddings for human-centric contents. We also design a simple-yet-effective MaskFusion module, which classifies and fuses multi-view features into 3D semantic masks without complex voting and grouping mechanisms. The design of decoupling mask proposals and text input also significantly boosts the efficiency of per-prompt inference. Experimental results on various 3D human datasets show that our method outperforms current state-of-the-art open-vocabulary 3D segmentation methods by a large margin. In addition, we show that our method can be directly applied to various 3D representations including meshes, point clouds, and 3D Gaussian Splatting.

The interest in matching non-rigidly deformed shapes represented as raw point clouds is rising due to the proliferation of low-cost 3D sensors. Yet, the task is challenging since point clouds are irregular and there is a lack of intrinsic shape information. We propose to tackle these challenges by learning a new shape representation -- a per-point high dimensional embedding, in an embedding space where semantically similar points share similar embeddings. The learned embedding has multiple beneficial properties: it is aware of the underlying shape geometry and is robust to shape deformations and various shape artefacts, such as noise and partiality. Consequently, this embedding can be directly employed to retrieve high-quality dense correspondences through a simple nearest neighbor search in the embedding space. Extensive experiments demonstrate new state-of-the-art results and robustness in numerous challenging non-rigid shape matching benchmarks and show its great potential in other shape analysis tasks, such as segmentation.

Current image-to-3D approaches suffer from high computational costs and lack scalability for high-resolution outputs. In contrast, we introduce a novel framework to directly generate explicit surface geometry and texture using multi-view 2D depth and RGB images along with 3D Gaussian features using a repurposed Stable Diffusion model. We introduce a depth branch into U-Net for efficient and high quality multi-view, cross-domain generation and incorporate epipolar attention into the latent-to-pixel decoder for pixel-level multi-view consistency depths. By back-projecting the generated depth pixels into 3D space, we create a structured 3D representation that can be either rendered via Gaussian splatting or extracted to high-quality meshes, thereby leveraging additional novel view synthesis loss to further improve our performance. Extensive experiments demonstrate that our method surpasses existing baselines in geometry and texture quality while achieving significantly faster generation time.

We propose MeshUp, a technique that deforms a 3D mesh towards multiple target concepts, and intuitively controls the region where each concept is expressed. Conveniently, the concepts can be defined as either text queries, e.g., “a dog” and “a turtle,” or inspirational images, and the local regions can be selected as any number of vertices on the mesh. We can effectively control the influence of the concepts and mix them together using a novel score distillation approach, referred to as the Blended Score Distillation (BSD). BSD operates on each attention layer of the denoising U-Net of a diffusion model as it extracts and injects the per-objective activations into a unified denoising pipeline from which the deformation gradients are calculated. To localize the expression of these activations, we create a probabilistic Region of Interest (ROI) map on the surface of the mesh, and turn it into 3D-consistent masks that we use to control the expression of these activations. We demonstrate the effectiveness of BSD empirically and show that it can deform various meshes towards multiple objectives.

White Light Interferometry (WLI) is a precise optical tool for measuring the 3D topography of microstructures. However, conventional WLI cannot capture the natural color of a sample's surface, which is essential for many microscale research applications that require both 3D geometry and color information. Previous methods have attempted to overcome this limitation by modifying WLI hardware and analysis software, but these solutions are often costly. In this work, we address this challenge from a computer vision multi-modal reconstruction perspective for the first time. We introduce OpticFusion, a novel approach that uses an additional digital optical microscope (OM) to achieve 3D reconstruction with natural color textures using multi-view WLI and OM images. Our method employs a two-step data association process to obtain the poses of WLI and OM data. By leveraging the neural implicit representation, we fuse multi-modal data and apply color decomposition technology to extract the sample's natural color. Tested on our multi-modal dataset of various microscale samples, OpticFusion achieves detailed 3D reconstructions with color textures. Our method provides an effective tool for practical applications across numerous microscale research fields. The source code and our real-world dataset are available at https://github.com/zju3dv/OpticFusion.

While text-to-3D and image-to-3D generation tasks have received considerable attention, one important but under-explored field between them is controllable text-to-3D generation, which we mainly focus on in this work. To address this task, 1) we introduce Multi-view ControlNet (MVControl), a novel neural network architecture designed to enhance existing pre-trained multi-view diffusion models by integrating additional input conditions, such as edge, depth, normal, and scribble maps. Our innovation lies in the introduction of a conditioning module that controls the base diffusion model using both local and global embeddings, which are computed from the input condition images and camera poses. Once trained, MVControl is able to offer 3D diffusion guidance for optimization-based 3D generation. And, 2) we propose an efficient multi-stage 3D generation pipeline that leverages the benefits of recent large reconstruction models and score distillation algorithm. Building upon our MVControl architecture, we employ a unique hybrid diffusion guidance method to direct the optimization process. In pursuit of efficiency, we adopt 3D Gaussians as our representation instead of the commonly used implicit representations. We also pioneer the use of SuGaR, a hybrid representation that binds Gaussians to mesh triangle faces. This approach alleviates the issue of poor geometry in 3D Gaussians and enables the direct sculpting of fine-grained geometry on the mesh. Extensive experiments demonstrate that our method achieves robust generalization and enables the controllable generation of high-quality 3D content.

In the pursuit of efficient automated content creation, procedural generation, leveraging modifiable parameters and rule-based systems, emerges as a promising approach. Nonetheless, it could be a demanding endeavor, given its intricate nature necessitating a deep understanding of rules, algorithms, and parameters. To reduce workload, we introduce 3D-GPT, a framework utilizing large language models~(LLMs) for instruction-driven 3D modeling. 3D-GPT positions LLMs as proficient problem solvers, dissecting the procedural 3D modeling tasks into accessible segments and appointing the apt agent for each task. 3D-GPT integrates three core agents: the task dispatch agent, the conceptualization agent, and the modeling agent. They collaboratively achieve two objectives. First, it enhances concise initial scene descriptions, evolving them into detailed forms while dynamically adapting the text based on subsequent instructions. Second, it integrates procedural generation, extracting parameter values from enriched text to effortlessly interface with 3D software for asset creation. Our empirical investigations confirm that 3D-GPT not only interprets and executes instructions, delivering reliable results but also collaborates effectively with human designers. Furthermore, it seamlessly integrates with Blender, unlocking expanded manipulation possibilities. Our work highlights the potential of LLMs in 3D modeling, offering a basic framework for future advancements in scene generation and animation.

In this work, we introduce FourieRF, a novel approach for achieving fast and high-quality reconstruction in the few-shot setting. Our method effectively parameterizes features through an explicit curriculum training procedure, incrementally increasing scene complexity during optimization. Experimental results show that the prior induced by our approach is both robust and adaptable across a wide variety of scenes, establishing FourieRF as a strong and versatile baseline for the few-shot rendering problem. While our approach significantly reduces artifacts, it may still lead to reconstruction errors in severely under-constrained scenarios, particularly where view occlusion leaves parts of the shape uncovered. In the future, our method could be enhanced by integrating foundation models to complete missing parts using large data-driven priors.

We introduce RealmDreamer, a technique for generating forward-facing 3D scenes from text descriptions. Our method optimizes a 3D Gaussian Splatting representation to match complex text prompts using pretrained diffusion models. Our key insight is to leverage 2D inpainting diffusion models conditioned on an initial scene estimate to provide low variance and high-fidelity estimates of unknown regions during 3D distillation. In conjunction, we imbue correct geometry with geometric distillation from a depth diffusion model, conditioned on samples from the inpainting model. We find that the initialization of the optimization is crucial, and provide a principled methodology for doing so. Notably, our technique doesn't require video or multi-view data and can synthesize various high-quality 3D scenes in different styles with complex layouts. Further, the generality of our method allows 3D synthesis from a single image. As measured by a comprehensive user study, our method outperforms all existing approaches, preferred by 88-95%. We encourage viewing the supplemental website and video. Project page: https://realmdreamer.github.io/

We introduce a new approach for reconstruction and novel view synthesis of unbounded real-world scenes. In contrast to previous methods using either volumetric fields, grid-based models, or discrete point cloud proxies, we propose a hybrid scene representation, which *implicitly* encodes the geometry in a continuous octree-based probability field and view-dependent appearance in a multi-resolution hash grid. This allows for extraction of arbitrary *explicit* point clouds, which can be rendered using rasterization. In doing so, we combine the benefits of both worlds and retain favorable behavior during optimization: Our novel implicit point cloud representation and differentiable bilinear rasterizer enable fast rendering while preserving the fine geometric detail captured by volumetric neural fields. Furthermore, this representation does not depend on priors like structure-from-motion point clouds. Our method achieves state-of-the-art image quality on common benchmarks. Furthermore, we achieve fast inference at interactive frame rates, and can convert our trained model into a large, explicit point cloud to further enhance performance.

Current text-to-avatar methods often rely on implicit representations (e.g., NeRF, SDF, and DMTet), leading to 3D content that artists cannot easily edit and animate in graphics software. This paper introduces a novel framework for generating stylized head avatars from text guidance, which leverages locally learnable mesh deformation and 2D diffusion priors to achieve high-quality digital assets for attribute-preserving manipulation. Given a template mesh, our method represents mesh deformation with per-face Jacobians and adaptively modulates local deformation using a learnable vector field. This vector field enables anisotropic scaling while preserving the rotation of vertices, which can better express identity and geometric details. We also employ landmark- and contour-based regularization terms to balance the expressiveness and plausibility of generated head avatars from multiple views without relying on any specific shape prior. Our framework can generate realistic shapes and textures that can be further edited via text, while supporting seamless editing using the preserved attributes from the template mesh, such as 3DMM parameters, blendshapes, and UV coordinates. Extensive experiments demonstrate that our framework can generate diverse and expressive head avatars with high-quality meshes that artists can easily manipulate in 3D graphics software, facilitating downstream applications such as efficient asset creation and animation with preserved attributes.

We introduce SPAFormer, an innovative model designed to overcome the combinatorial explosion challenge in the 3D Part Assembly (3D-PA) task. This task requires accurate prediction of each part's poses in sequential steps. As the number of parts increases, the possible assembly combinations increase exponentially, leading to a combinatorial explosion that severely hinders the efficacy of 3D-PA. SPAFormer addresses this problem by leveraging weak constraints from assembly sequences, effectively reducing the solution space's complexity. Since the sequence of parts conveys construction rules similar to sentences structured through words, our model explores both parallel and autoregressive generation. We further strengthen SPAFormer through knowledge enhancement strategies that utilize the attributes of parts and their sequence information, enabling it to capture the inherent assembly pattern and relationships among sequentially ordered parts. We also construct a more challenging benchmark named PartNet-Assembly covering 21 varied categories to more comprehensively validate the effectiveness of SPAFormer. Extensive experiments demonstrate the superior generalization capabilities of SPAFormer, particularly with multi-tasking and in scenarios requiring long-horizon assembly. Code is available at https://github.com/xuboshen/SPAFormer.

Generative modeling of 3D human bodies have been studied extensively in computer vision. The core is to design a compact latent representation that is both expressive and semantically interpretable, yet existing approaches struggle to achieve both requirements. In this work, we introduce JADE, a generative framework that learns the variations of human shapes with fined-grained control. Our key insight is a joint-aware latent representation that decomposes human bodies into skeleton structures, modeled by joint positions, and local surface geometries, characterized by features attached to each joint. This disentangled latent space design enables geometric and semantic interpretation, facilitating users with flexible controllability. To generate coherent and plausible human shapes under our proposed decomposition, we also present a cascaded pipeline where two diffusions are employed to model the distribution of skeleton structures and local surface geometries respectively. Extensive experiments are conducted on public datasets, where we demonstrate the effectiveness of JADE framework in multiple tasks in terms of autoencoding reconstruction accuracy, editing controllability and generation quality compared with existing methods.

The ability to cope with out-of-distribution (OOD) corruptions and adversarial attacks is crucial in real-world safety-demanding applications. In this study, we develop a general mechanism to increase point clouds neural networks robustness based on focus analysis. Recent studies have revealed the phenomenon of \textit{Overfocusing}, which leads to a performance drop. When the network is primarily influenced by small input regions, it becomes less robust and prone to misclassify under noise and corruptions. However, quantifying overfocusing is still vague and lacks clear definitions. Here, we provide a mathematical definition of \textbf{focus}, \textbf{overfocusing} and \textbf{underfocusing}. The notions are general, but in this study, we specifically investigate the case of 3D point clouds. We observe that corrupted sets result in a biased focus distribution compared to the clean training set. We show that as focus distribution deviates from the one learned in the training phase - classification performance deteriorates. We thus propose a parameter-free \textbf{refocusing} algorithm that aims to unify all corruptions under the same distribution. We validate our findings on a 3D zero-shot classification task, achieving SOTA in robust 3D classification on ModelNet-C dataset, and in adversarial defense against Shape-Invariant attack.

We introduce Gaussian Garments, a novel approach for reconstructing realistic-looking, simulation-ready, garments from multi-view videos. Our method represents garments with a combination of a 3D mesh and a Gaussian texture that encodes both the color and high-frequency surface texture. With this, it is able to model rich textures and surface detail including complex materials such as fur. As part of our reconstruction process, we demonstrate how such a representation can be used to register a 3D mesh to multi-view videos. Our reconstructed Gaussian Garments can be animated on unseen body shapes and motions using physics-based simulation. We show how a learned GNN-based physical simulator can be used to optimize the garment's material parameters to match its real behavior. Finally, we devise a procedure for automatically ordering virtual garments, allowing us to combine the captured individual garments into multi-garment outfits.

Implicit neural fields have made remarkable progress in reconstructing 3D surfaces from multiple images; however, they encounter challenges when it comes to separating individual objects within a scene. Previous approaches to this problem require ground-truth segmentation masks and introduce floating artifacts in occluded parts of the scene. We address these challenges with ObjectCarver. ObjectCarver requires no ground-truth segmentation; all it needs is just a few user clicks in a single view. ObjectCarver also introduces a new loss function that prevents floaters and avoids inappropriate carving-out due to occlusion. Finally, ObjectCarver uses a simple initialization technique that significantly speeds up the process while preserving geometric details. We demonstrate qualitatively and quantitatively on multiple datasets (including a new dataset and benchmark with complete ground-truth) that ObjectCarver produces more accurate reconstructions of each object while minimizing artifacts

Human Mesh Recovery (HMR) is an important yet challenging problem with applications across various domains including motion capture, augmented reality, and biomechanics. Accurately predicting human pose parameters from a single image remains a challenging 3D computer vision task. In this work, we introduce DeforHMR, a novel regression-based monocular HMR framework designed to enhance the prediction of human pose parameters using deformable attention transformers. DeforHMR leverages a novel query-agnostic deformable cross-attention mechanism within the transformer decoder to effectively regress the visual features extracted from a frozen pretrained vision transformer (ViT) encoder. The proposed deformable cross-attention mechanism allows the model to attend to relevant spatial features more flexibly and in a data-dependent manner. Equipped with a transformer decoder capable of spatially-nuanced attention, DeforHMR achieves state-of-the-art performance for single-frame regression-based methods on the widely used 3D HMR benchmarks 3DPW and RICH. By pushing the boundary on the field of 3D human mesh recovery through deformable attention, we introduce an new, effective paradigm for decoding local spatial information from large pretrained vision encoders in computer vision.

As humans move around, performing their daily tasks, they are able to recall where they have positioned objects in their environment, even if these objects are currently out of their sight. In this paper, we aim to mimic this spatial cognition ability. We thus formulate the task of *Out of Sight, Not Out of Mind* -- 3D tracking active objects using observations captured through an egocentric camera. We introduce a simple but effective approach to address this challenging problem, called Lift, Match, and Keep (LMK). LMK **lifts** partial 2D observations to 3D world coordinates, **matches** them over time using visual appearance, 3D location and interactions to form object tracks, and **keeps** these object tracks even when they go out-of-view of the camera. We benchmark LMK on 100 long videos from EPIC-KITCHENS. Our results demonstrate that spatial cognition is critical for correctly locating objects over short and long time scales. E.g., for one long egocentric video, we estimate the 3D location of 50 active objects. After 120 seconds, 57\% of the objects are correctly localized by LMK, compared to just 33\% by a recent 3D method for egocentric videos and 17\% by a general 2D tracking method.