CAdam reinterprets densification in generative 3DGS as signal verification via gradient-moment interference, quantile context, and SNR gating to achieve large reductions in primitive count with comparable quality.
hub Canonical reference
Hunyuan3D 2.5: Towards High-Fidelity 3D Assets Generation with Ultimate Details
Canonical reference. 90% of citing Pith papers cite this work as background.
40
Pith papers citing it
Background
90% of classified citations
abstract
In this report, we present Hunyuan3D 2.5, a robust suite of 3D diffusion models aimed at generating high-fidelity and detailed textured 3D assets. Hunyuan3D 2.5 follows two-stages pipeline of its previous version Hunyuan3D 2.0, while demonstrating substantial advancements in both shape and texture generation. In terms of shape generation, we introduce a new shape foundation model -- LATTICE, which is trained with scaled high-quality datasets, model-size, and compute. Our largest model reaches 10B parameters and generates sharp and detailed 3D shape with precise image-3D following while keeping mesh surface clean and smooth, significantly closing the gap between generated and handcrafted 3D shapes. In terms of texture generation, it is upgraded with phyiscal-based rendering (PBR) via a novel multi-view architecture extended from Hunyuan3D 2.0 Paint model. Our extensive evaluation shows that Hunyuan3D 2.5 significantly outperforms previous methods in both shape and end-to-end texture generation.
hub tools
citation-role summary
background 9
baseline 1
citation-polarity summary
years
2026 40representative citing papers
VS3D performs local 3D asset editing by injecting reconstruction-anchored source signals, partial-mean guidance, and twin-agreement residuals into the velocity sampler to control edit strength and preserve identity.
THOM is a training-free two-stage framework that generates physically plausible hand-object 3D meshes directly from text by combining text-guided Gaussians with contact-aware physics optimization and VLM refinement.
ATATA enables fast joint inference of structurally aligned pairs using Rectified Flow models via segment transport, improving state-of-the-art for image and video generation while matching 3D quality at much higher speed.
An end-to-end 3D editing framework achieves high-fidelity local edits from coarse bounding boxes and 2D image prompts using region-aware loss reweighting and a large-scale parts-derived training dataset.
PixGS is a single-stage pixel-space diffusion model that directly produces high-quality 3D Gaussian Splats from text or images in ~1s, outperforming multi-stage latent methods on standard benchmarks.
An automated real-to-sim pipeline builds digital twins and affordance-preserving cousins from video, yielding sim evaluations that correlate with real robot policy success and zero-shot sim-to-real gains.
HiFiVe is a training-free framework using an auto-regressive texture refinement pipeline with depth-based warping, multi-view fusion, and symmetry to enhance both texture and geometry fidelity in vehicle generation from 2D priors.
T2Mo generates controllable dynamic 3D shapes by conditioning on both text semantics and 3D trajectories with a shape-grounded embedding for arbitrary inputs.
Adaptive saliency-guided supervoxel tokenization cuts 3D AR token length to 12.8% of uniform voxels while claiming SOTA quality and ~10× speedup on Trellis-500K.
FoundObj uses foundation-model priors as RL rewards to discover multi-class 3D objects from point clouds without scene-level labels.
Helix4D generates high-quality dynamic 4D meshes from videos by extending Trellis2 with sliding-window cross-frame attention anchored on the first frame and a repurposed 4D temporal encoding.
OMGTex is a geometry-free diffusion pipeline for multi-style facial UV texture reconstruction that adds gradient-guided refinement and semantic training to enable editing, plus a new paired dataset CANVAS.
ROAR-3D adds a token-wise view router and dual-stream attention to pretrained single-view 3D generators so they can use arbitrary unposed images for higher-fidelity output.
R-DMesh proposes a VAE-based disentanglement of base mesh, motion trajectories, and rectification offset plus Triflow Attention and rectified-flow diffusion to produce 4D meshes aligned to video despite initial pose mismatch.
A new framework combines self-attention on the Oblique manifold with bidirectional geodesic cross-attention on the Lorentz hyperboloid to improve both localization accuracy and descriptive coherence in 3D dense captioning.
DVD applies discrete diffusion directly to voxel occupancy for 3D generation, uncertainty estimation via entropy, and single-round editing via block perturbation fine-tuning.
VISER is a new visually realistic simulation benchmark for robot manipulation tasks that uses PBR materials and MLLM-assisted asset generation, achieving 0.92 Pearson correlation with real-world policy performance.
A single-image head reconstruction method uses coarse-to-fine optimization with normal consistency, landmarks, and geometry-aware constraints on curvature and conformality to produce meshes with industry-grade topology and preserved facial identity.
An animator-centric skeleton generation method that uses semantic-aware tokenization and a learnable density interval module to produce controllable, high-quality skeletons on complex 3D meshes.
BVE framework enables text-guided 3D editing beyond voxel limits by combining self-constructed data, lightweight semantic injection, and annotation-free masking to preserve local invariance.
GraG reconstructs dynamic 3D hand-object interactions from monocular video 6.4x faster than prior work by using compact Sum-of-Gaussians tracking initialized from large models and refined with 2D losses.
Pair2Scene generates complex 3D scenes beyond training data by training a network on local object-pair placement rules and applying them recursively with collision-aware sampling.
DailyArt recovers full joint parameters of articulated objects from a single static image by synthesizing an opened state and comparing discrepancies, supporting downstream part-level novel state synthesis.
citing papers explorer
-
CAdam: Context-Adaptive Moment Estimation for 3D Gaussian Densification in Generative Distillation
CAdam reinterprets densification in generative 3DGS as signal verification via gradient-moment interference, quantile context, and SNR gating to achieve large reductions in primitive count with comparable quality.
-
Velocity-Space 3D Asset Editing
VS3D performs local 3D asset editing by injecting reconstruction-anchored source signals, partial-mean guidance, and twin-agreement residuals into the velocity sampler to control edit strength and preserve identity.
-
THOM: Generating Physically Plausible Hand-Object Meshes From Text
THOM is a training-free two-stage framework that generates physically plausible hand-object 3D meshes directly from text by combining text-guided Gaussians with contact-aware physics optimization and VLM refinement.
-
ATATA: One Algorithm to Align Them All
ATATA enables fast joint inference of structurally aligned pairs using Rectified Flow models via segment transport, improving state-of-the-art for image and video generation while matching 3D quality at much higher speed.
-
EditVerse3D: High-Quality 3D Object Editing with Region-Aware Learning
An end-to-end 3D editing framework achieves high-fidelity local edits from coarse bounding boxes and 2D image prompts using region-aware loss reweighting and a large-scale parts-derived training dataset.
-
PixGS: Pixel-Space Diffusion for Direct 3D Gaussian Splat Generation
PixGS is a single-stage pixel-space diffusion model that directly produces high-quality 3D Gaussian Splats from text or images in ~1s, outperforming multi-stage latent methods on standard benchmarks.
-
SimFoundry: Modular and Automated Scene Generation for Policy Learning and Evaluation
An automated real-to-sim pipeline builds digital twins and affordance-preserving cousins from video, yielding sim evaluations that correlate with real robot policy success and zero-shot sim-to-real gains.
-
HiFiVe: High-Fidelity Vehicle Generation Leveraging Auto-Regressive 2D Generative Priors
HiFiVe is a training-free framework using an auto-regressive texture refinement pipeline with depth-based warping, multi-view fusion, and symmetry to enhance both texture and geometry fidelity in vehicle generation from 2D priors.
-
Controllable Dynamic 3D Shape Generation via 3D Trajectories and Text
T2Mo generates controllable dynamic 3D shapes by conditioning on both text semantics and 3D trajectories with a shape-grounded embedding for arbitrary inputs.
-
SuperVoxelGPT: Adaptive and Ordered 3D Tokenization for Autoregressive Shape Generation
Adaptive saliency-guided supervoxel tokenization cuts 3D AR token length to 12.8% of uniform voxels while claiming SOTA quality and ~10× speedup on Trellis-500K.
-
FoundObj: Self-supervised Foundation Models as Rewards for Label-free 3D Object Segmentation
FoundObj uses foundation-model priors as RL rewards to discover multi-class 3D objects from point clouds without scene-level labels.
-
Helix4D: Complex 4D Mesh Generation
Helix4D generates high-quality dynamic 4D meshes from videos by extending Trellis2 with sliding-window cross-frame attention anchored on the first frame and a repurposed 4D temporal encoding.
-
OMGTex: One-stage Multi-style Facial Texture Reconstruction without Geometry Guidance
OMGTex is a geometry-free diffusion pipeline for multi-style facial UV texture reconstruction that adds gradient-guided refinement and semantic training to enable editing, plus a new paired dataset CANVAS.
-
ROAR-3D: Routing Arbitrary Views for High-Fidelity 3D Generation
ROAR-3D adds a token-wise view router and dual-stream attention to pretrained single-view 3D generators so they can use arbitrary unposed images for higher-fidelity output.
-
R-DMesh: Video-Guided 3D Animation via Rectified Dynamic Mesh Flow
R-DMesh proposes a VAE-based disentanglement of base mesh, motion trajectories, and rectification offset plus Triflow Attention and rectified-flow diffusion to produce 4D meshes aligned to video despite initial pose mismatch.
-
Curvature-Aware Captioning:Leveraging Geodesic Attention for 3D Scene Understanding
A new framework combines self-attention on the Oblique manifold with bidirectional geodesic cross-attention on the Lorentz hyperboloid to improve both localization accuracy and descriptive coherence in 3D dense captioning.
-
DVD: Discrete Voxel Diffusion for 3D Generation and Editing
DVD applies discrete diffusion directly to voxel occupancy for 3D generation, uncertainty estimation via entropy, and single-round editing via block perturbation fine-tuning.
-
Toward Visually Realistic Simulation: A Benchmark for Evaluating Robot Manipulation in Simulation
VISER is a new visually realistic simulation benchmark for robot manipulation tasks that uses PBR materials and MLLM-assisted asset generation, achieving 0.92 Pearson correlation with real-world policy performance.
-
High-Fidelity Single-Image Head Modeling with Industry-Grade Topology
A single-image head reconstruction method uses coarse-to-fine optimization with normal consistency, landmarks, and geometry-aware constraints on curvature and conformality to produce meshes with industry-grade topology and preserved facial identity.
-
Animator-Centric Skeleton Generation on Objects with Fine-Grained Details
An animator-centric skeleton generation method that uses semantic-aware tokenization and a learnable density interval module to produce controllable, high-quality skeletons on complex 3D meshes.
-
Beyond Voxel 3D Editing: Learning from 3D Masks and Self-Constructed Data
BVE framework enables text-guided 3D editing beyond voxel limits by combining self-constructed data, lightweight semantic injection, and annotation-free masking to preserve local invariance.
-
Grasp in Gaussians: Fast Monocular Reconstruction of Dynamic Hand-Object Interactions
GraG reconstructs dynamic 3D hand-object interactions from monocular video 6.4x faster than prior work by using compact Sum-of-Gaussians tracking initialized from large models and refined with 2D losses.
-
Pair2Scene: Learning Local Object Relations for Procedural Scene Generation
Pair2Scene generates complex 3D scenes beyond training data by training a network on local object-pair placement rules and applying them recursively with collision-aware sampling.
-
DailyArt: Discovering Articulation from Single Static Images via Latent Dynamics
DailyArt recovers full joint parameters of articulated objects from a single static image by synthesizing an opened state and comparing discrepancies, supporting downstream part-level novel state synthesis.
-
UniRecGen: Unifying Multi-View 3D Reconstruction and Generation
UniRecGen unifies reconstruction and generation via shared canonical space and disentangled cooperative learning to produce complete, consistent 3D models from sparse views.
-
StoryBlender: Inter-Shot Consistent and Editable 3D Storyboard with Spatial-temporal Dynamics
StoryBlender generates inter-shot consistent editable 3D storyboards using a three-stage pipeline of semantic-spatial grounding, canonical asset materialization, and spatial-temporal dynamics with agent-based verification.
-
DreamCharacter-1: From 3D Generative Foundation Models to Product-Ready Character Generation
A post-adaptation pipeline on 3D foundation models yields higher-fidelity, faster, animation-ready single-image 3D characters than prior open methods on reported benchmarks.
-
MM-TRELLIS: Point-Cloud Guided Multi-Modal 3D Vehicle Generation in Autonomous Driving
MM-TRELLIS extends TRELLIS with LiDAR point-cloud guidance and multi-view image conditioning plus voxel filtering to generate high-fidelity 3D vehicle meshes from in-the-wild driving data.
-
ISAP-3D: Identity-Slot Aligned Part-Aware 3D Generation
ISAP-3D proposes identity-slot aligned modeling with semantic identity tokens and one-to-one layout prediction to achieve stable part-aware 3D generation.
-
AnimaSpark: A Feed-Forward Method for Animating Arbitrary 3D Objects
AnimaSpark is a feed-forward pipeline that renders rigged 3D models to multi-layer images, generates video motion, tracks projected keypoints, and lifts 2D planar transforms into 3D skeletal animation.
-
ARAPDiffusion: ARAP Regularization for Diffusion-Based Deformable Shape Space Learning
ARAPDiffusion adds ARAP regularization losses to a latent diffusion model and alternates between improving the shape autoencoder and the diffusion model to learn continuous deformable shape spaces from limited data.
-
WorldAct: Activating Monolithic 3D Worlds into Interactive-Ready Object-Centric Scenes
WorldAct activates monolithic 3D worlds into interactive scenes via multimodal agent-guided decomposition, geometrically aligned mesh reconstruction, and 3D inpainting.
-
DataEvolver: Let Your Data Build and Improve Itself via Goal-Driven Loop Agents
DataEvolver introduces a reusable framework with generation-time self-correction and validation-time self-expansion loops that improves visual datasets, shown to outperform baselines on an object-rotation task.
-
Asset Harvester: Extracting 3D Assets from Autonomous Driving Logs for Simulation
Asset Harvester converts sparse in-the-wild object observations from AV driving logs into complete simulation-ready 3D assets via data curation, geometry-aware preprocessing, and a SparseViewDiT model that couples sparse-view multiview generation with 3D Gaussian lifting.
-
Hitem3D 2.0: Multi-View Guided Native 3D Texture Generation
Hitem3D 2.0 combines multi-view image synthesis with native 3D texture projection to improve completeness, cross-view consistency, and geometry alignment over prior methods.
-
FlowObject: Flow Steering for Bridging Generative Priors and Reconstruction Fidelity
FlowObject reformulates sparse-view 3D reconstruction as a training-free guided inverse problem in flow-matching models, augmented by 3DGS refinement to improve geometric completeness and fidelity.
-
ABot-Earth 0.5: Generative 3D Earth Model
ABot-Earth 0.5 is a 3DGS-based generative model trained on real-world urban reconstructions that synthesizes novel 3D scenes from satellite imagery in under 10 minutes per square kilometer.
-
IsaacIPC: Coupling High-Fidelity Simulation and Realistic Rendering for Contact-Rich Robotic Systems
IsaacIPC couples IPC contact simulation with IsaacSim/Lab for realistic rendering and introduces GMCP for tactile contact pressure modeling, demonstrated on robotic examples.
- EmbodiedGen V2: An Agentic, Simulation-Ready 3D World Engine for Embodied AI
- BrickAnything: Geometry-Conditioned Buildable Brick Generation with Structure-Aware Tokenization