Skip to main content
GameDev.net gamedev.net

Procedural Content News

The latest Procedural Content coverage curated for game developers.

Mira-Scene: Pixel-Aligned Layouts for Generative 3D Scene Reconstruction

arXiv cs.GR details Mira-Scene, a 3D scene reconstruction method that aligns generated objects to pixels instead of relying on sparse pose guesses. For game teams building scene tools or AI-assisted …

Research arXiv cs.GR · 3 days, 22 hours ago
Physically Based Rendering in the Latent Space

arXiv cs.GR details a method that pushes physically based rendering into the latent space used by diffusion models. The approach aims to make image generation more controllable by simulating light …

Research arXiv cs.GR · 4 days, 22 hours ago
PART: Learning 3D Part Assembly and Retrieval with Transformers

arXiv cs.GR details PART, a transformer-based system that can retrieve 3D parts from a library and assemble them into a target shape. The approach tackles variable part counts and 6-DoF …

Research arXiv cs.GR · 1 week ago
KaiNinja: Extending Native 3D Generators to the Part Level

KaiNinja extends native 3D generation from whole objects to editable parts, targeting the pain point of fused meshes that are awkward to rig, simulate, or modify. Built on TRELLIS.2, it …

Research arXiv cs.GR · 1 week, 3 days ago
Articulate That Object Part (ATOP): 3D Part Articulation from Text and via Motion Personalization

ATOP is a new 3D part-articulation method that lets creators drive object motion from text and personalize it with motion examples. For game teams, that points to faster iteration on …

Research ACM Graphics · 1 week, 4 days ago
Grow3D: Hierarchical Next-Scale Octree Prediction for Fast and High-Fidelity 3D Shape Generation

Researchers have introduced Grow3D, a hierarchical octree-based method for generating 3D shapes with higher fidelity and faster inference. The approach predicts the next scale of an octree step by step, …

Research ACM Graphics · 1 week, 4 days ago
SNAP3D: Physically Grounded 3D Parts for Assembly from a Single Image

SNAP3D tackles a common failure mode in part-aware 3D generation: parts that look right but cannot actually assemble. The system uses physics feedback to reduce interpenetration, infer contact relationships, and …

Research arXiv cs.GR · 1 week, 4 days ago
Evaluating Constrained Iterative Refinement for Scalable Vector Graphics Generation with Off-the-Shelf VLMs

New research shows off-the-shelf vision-language models can be pushed toward SVG generation with constrained decoding and iterative refinement. The biggest win is higher compilation success, but the models still struggle …

Research arXiv cs.GR · 3 weeks, 3 days ago
MultiCube: Compositional 3D Generation With Part-Level Semantic and Spatial Control

arXiv cs.GR details MultiCube, a 3D generation method that gives creators part-level control over both meaning and placement. It builds compositional meshes from a text prompt, a part schema, and …

Research arXiv cs.GR · 1 month ago
MeshFlow: Mesh Generation with Equivariant Flow Matching

MeshFlow generates triangle meshes directly with equivariant flow matching, sidestepping long autoregressive mesh sequences. The method respects face and vertex permutation symmetries and reports quality on par with leading mesh …

Research arXiv cs.GR · 1 month ago
aDSL: Agentic 3D Creation via Joint Agent-Program Design

A new 3D creation pipeline pairs a custom DSL with a multi-agent LLM workflow to make programmatic content generation less brittle. The system replaces fragile absolute coordinates with relational operators, …

Research arXiv cs.GR · 1 month ago
MegaParts: Scaling Part-Aware 3D Object Generation to 300 Parts via Token-Efficient Autoregressive Modeling

MegaParts pushes part-aware 3D generation to 300 parts by pairing an autoregressive model with a token-efficient VQ shape tokenizer. For game teams, the practical angle is finer control over complex …

Research arXiv cs.GR · 1 month, 1 week ago
RealMat: Realistic Materials with Diffusion and Reinforcement Learning

RealMat combines Stable Diffusion XL with reinforcement learning to generate more believable material maps for 3D authoring. The pipeline starts from synthetic 2×2 material grids, then pushes the model toward …

Research arXiv cs.GR · 1 month, 1 week ago
Meshy T2: Fast Native Mesh Generation with Flow Matching

Meshy T2 can turn an image into a mesh in a median of 6 seconds, using flow matching instead of slow autoregressive token decoding. The system generates topology natively, preserving …

Research arXiv cs.GR · 1 month, 3 weeks ago
Elastic Curves via Geometric Mechanics

A new geometric-mechanics formulation reframes elastic curves as critical points of length with fixed area and volume vectors, which is a cleaner fit for rigid-body symmetries. For graphics and simulation …

Research arXiv cs.GR · 1 month, 3 weeks ago
Nova3D: Code-Native Generation of Programmable 3D Assets

Nova3D turns 3D asset generation into executable Blender code instead of a sealed mesh. The system produced valid outputs for all 54 benchmark items, with named parts, assembly trees, numeric …

Research arXiv cs.GR · 1 month, 4 weeks ago
Spatula: Exploring On-Demand In-Situ Interfaces and Interaction for Attribute Control

A new HCI/graphics prototype called Spatula generates on-demand, in-situ controls for motion graphics attributes, aiming to make iterative tuning feel closer to direct manipulation. In a 12-person study, it was …

Research arXiv cs.GR · 2 months, 1 week ago
3D-CBM: A Framework for Concept-Based Interpretability in Generative 3D Modeling

A new 3D generative modeling framework uses concept bottlenecks to make geometry models more interpretable and steerable. In a proof-of-concept part-manipulation test, it reached 88.8% concept prediction accuracy and a …

Research arXiv cs.GR · 3 months, 2 weeks ago
MeshFlow: Efficient Artistic Mesh Generation via MeshVAE and Flow-based Diffusion Transformer

MeshFlow replaces slow auto-regressive mesh tokenization with a continuous latent VAE plus a rectified-flow transformer, and it claims 18x faster generation than the fastest AR baseline. For graphics and tools …

Research arXiv cs.GR · 3 months, 3 weeks ago
CompoSE: Compositional Synthesis and Editing of 3D Shapes via Part-Aware Control

CompoSE is a new 3D shape synthesis system that lets creators place coarse part layouts and get back editable, part-separated objects. It uses a diffusion transformer to balance local part …

Research arXiv cs.GR · 4 months ago
Page 1 of 2 Next