archive
Every paper Pith has read. Search by title, abstract, or pith.
456 papers in cs.GR · page 6
-
Neural networks compress temporal lightmaps for dynamic GI
Neural Dynamic GI: Random-Access Neural Compression for Temporal Lightmaps in Dynamic Lighting Environments
-
Neural compression trims lightmap storage for dynamic lighting
Neural Dynamic GI: Random-Access Neural Compression for Temporal Lightmaps in Dynamic Lighting Environments
-
Transformer converts volumes to Gaussians in one forward pass
VVGT: Visual Volume-Grounded Transformer
-
Integer twists on edges turn meshes into linked knot structures
Twisted Edges: A Unified Framework for Designing Linked Knot (LK) Structures Using Labeled Non-Manifold Surface Meshes
-
Multi-modal LLM predicts MOOC satisfaction more accurately than text alone
Predicting User Satisfaction in Online Education Platforms: A Large Language Model Based Multi-Modal Review Mining Framework
-
INR features enable accurate ROI classification with sparse labels
NeuVolEx: Implicit Neural Features for Volume Exploration
-
Modified Harris cuts corner detection time for expression analysis
Product Review Based on Optimized Facial Expression Detection
-
Complex-valued method upsamples holograms while keeping linear depth
CV-HoloSR: Hologram to hologram super-resolution through volume-upsampling three-dimensional scenes
-
Conducting gestures reduced to cubic segments and quintic timing
A Minimal Mathematical Model for Conducting Patterns
-
NumPy batch transfer and Numba JIT enable fast Python game scripting
Infernux: A Python-Native Game Engine with JIT-Accelerated Scripting
-
One pass creates animatable human Gaussians from images
Real-Time Human Reconstruction and Animation using Feed-Forward Gaussian Splatting
-
Chip renders 3D Gaussian Splatting at 129 FPS in full HD
A 129FPS Full HD Real-Time Accelerator for 3D Gaussian Splatting
-
Latent fusion adds fatigue to motions without fatigue data
FatigueFusion: Latent Space Fusion for Fatigue-Driven Motion Synthesis
-
MRI change vectors sharpen epilepsy surgery forecasts
Neuro-Oracle: A Trajectory-Aware Agentic RAG Framework for Interpretable Epilepsy Surgical Prognosis
-
Alignment loss fixes grid errors in facade detection
Beyond Segmentation: Structurally Informed Facade Parsing from Imperfect Images
-
Visual designs cut OTC drug label reading time
Enhance Comprehension of Over-the-Counter Drug Instructions for the General Public and Medical Professionals through Visualization Design
-
Strips as tokens generate artist meshes with native UV boundaries
Strips as Tokens: Artist Mesh Generation with Native UV Segmentation
-
MeshOn places accessories on 3D models without any surface intersections
MeshOn: Intersection-Free Mesh-to-Mesh Composition
-
One image produces a rigged 3D asset ready to animate
AniGen: Unified $S^3$ Fields for Animatable 3D Asset Generation
-
Skelebones rig improves reanimation of unseen poses by 17% PSNR
GaussiAnimate: Reconstruct and Rig Animatable Categories with Level of Dynamics
-
Dataset of 1.13M images lets virtual try-on show real garment fit
FIT: A Large-Scale Dataset for Fit-Aware Virtual Try-On
-
Ergonomic rules steer AI apartment layouts to higher comfort scores
What a Comfortable World: Ergonomic Principles Guided Apartment Layout Generation
-
Chart choice affects accuracy on distribution tasks
Investigating Performance and Practices with Univariate Distribution Charts
-
Progressive experts enable stable tracking of contact-rich character motions
Physics-Based Motion Tracking of Contact-Rich Interacting Characters
-
Non-reference net selects lowest indistinguishable resolution
Seeing enough: non-reference perceptual resolution selection for power-efficient client-side rendering
-
Non-reference model selects lowest indistinguishable video resolution
Seeing enough: non-reference perceptual resolution selection for power-efficient client-side rendering
-
Deforms 3D meshes to image styles while keeping topology
Image-Guided Geometric Stylization of 3D Meshes
-
Alternating refinement stabilizes articulated object modeling
GEAR: GEometry-motion Alternating Refinement for Articulated Object Modeling with Gaussian Splatting
-
Elastic prior stabilizes fast adaptation for long 4D sequences
Fast Spatial Memory with Elastic Test-Time Training
-
MoRight lets users control objects and cameras separately in videos
MoRight: Motion Control Done Right
-
MoZoo turns coarse animal meshes into realistic fur videos
MoZoo:Unleashing Video Diffusion power in animal fur and muscle simulation
-
Emulation maps energy costs of real-time 3D splatting on weak GPUs
Splats under Pressure: Exploring Performance-Energy Trade-offs in Real-Time 3D Gaussian Splatting under Constrained GPU Budgets
-
Model generates poster layouts from product images under user constraints
Image-aware Layout Generation with User Constraints for Poster Design
-
DesigNet adds designer continuity and alignment to AI vector graphics
DesigNet: Learning to Draw Vector Graphics as Designers Do
-
Deformable Gaussians separate simulation from viz parameters
GS-Surrogate: Deformable Gaussian Splatting for Parameter Space Exploration of Ensemble Simulations
-
Video diffusion inpaints realistic HDR into LDR footage
DiffHDR: Re-Exposing LDR Videos with Video Diffusion Models
-
Hair strands from monocular video match top quality at 10x speed
EfficientMonoHair: Fast Strand-Level Reconstruction from Monocular Video via Multi-View Direction Fusion
-
Small LLMs beat larger models at CAD-CAE design optimization
COSMO-Agent: Tool-Augmented Agent for Closed-loop Optimization,Simulation,and Modeling Orchestration
-
Simulated pedestrians now reason about scenes and consequences
CrowdVLA: Embodied Vision-Language-Action Agents for Context-Aware Crowd Simulation
-
Active set method accelerates transparent 3DGS
SparseOIT: Improving Order-Independent Transparency 3DGS via Active Set Method
-
Impulse-based assistance tracks exaggerated character motions
Neural Assistive Impulses: Synthesizing Exaggerated Motions for Physics-based Characters
-
On-device clicks query real objects in XR without cloud uploads
ClickAIXR: On-Device Multimodal Vision-Language Interaction with Real-World Objects in Extended Reality
-
Neural network generates six-way lightmaps on the fly for dynamic smoke
Real-time Neural Six-way Lightmaps
-
Splatting separates geometry from appearance in driving scenes
SpectralSplat: Appearance-Disentangled Feed-Forward Gaussian Splatting for Driving Scenes
-
SASAV is a proposed multi-agent AI system that autonomously profiles scientific data
SASAV: Self-Directed Agent for Scientific Analysis and Visualization
-
Streaming 3D Gaussians improves viewpoint flexibility over video
Streaming Real-Time Rendered Scenes as 3D Gaussians
-
Point tracking repositions 3D Gaussians to fix fast-motion artifacts
TrackerSplat: Exploiting Point Tracking for Fast and Robust Dynamic 3D Gaussians Reconstruction
-
Pretraining on 1M videos yields high-fidelity 3D avatars
Large-scale Codec Avatars: The Unreasonable Effectiveness of Large-scale Avatar Pretraining
-
3DGS blending reformulated for Tensor Cores yields 1.42x speedup
GEMM-GS: Accelerating 3D Gaussian Splatting on Tensor Cores with GEMM-Compatible Blending
-
Harmonic textures raise primitive 3D models to high-frequency quality
Neural Harmonic Textures for High-Quality Primitive Based Neural Reconstruction