GATOR: Generative and Agentic 3D Object Reconstruction From Casual Images
Summary
NVIDIA researchers present GATOR, a generative and agentic framework that reconstructs complete, textured 3D object assets and their scene-relative pose from one or more casual images, using a local modality mixer, text-guided semantic conditioning, and an edit-render-review agent loop for refinement.
View Cached Full Text
Cached at: 10/12/26, 06:41 PM
Paper page - GATOR: Generative and Agentic 3D Object Reconstruction From Casual Images
Source: https://huggingface.co/papers/2610.11215
Abstract
Reconstructingcomplete,scene-aligned3Dobjectsfromcasualimagesrequiresintegratingsparse,uncertainobservationsandinferringsurfaceshiddenbyocclusions.WepresentGATOR,agenerativeandagenticframeworkthatrecoverstexturedobjectassetsandtheirscene-relativeposefromoneormoreimages.Ourlocalmodalitymixercouplespatch-alignedRGB,target-mask,andpointmapfeaturesbeforecross-viewreasoning,preservingscenecontextwhiledistinguishingthetargetfromitssurroundings.Text-guidedsemanticconditioningcomplementsthesespatialcueswithcategorynamesandobjectcaptionsthroughstage-specificadaptersforstructure,geometry,andappearancegeneration.Thegeneratedassetinitializesamultimodalagent,providinginstance-specificgeometryandposefortargetedstructuralandtexturerefinementthroughanobservation-guidededit-render-reviewloop.Acrosssyntheticobjects,clutteredtabletops,andindoorscenes,GATORachievesstronggeometricandappearancefidelitywhilerecoveringscene-relativeposefromsparseobservations.Time-budgetcomparisonsandscene-levelsimulationfurtherdemonstratethereconstructionefficiencyandsimulationreadiness.Projectpage:https://research.nvidia.com/labs/lpr/gator/
View arXiv pageView PDFProject pageAdd to collection
Get this paper in your agent:
hf papers read 2610\.11215
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2610.11215 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2610.11215 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2610.11215 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
GenRecon: Bridging Generative Priors for Multi-View 3D Scene Reconstruction
GenRecon introduces a method for 3D scene reconstruction that integrates generative 3D priors with multi-view image conditioning, achieving high-fidelity, editable mesh reconstructions of indoor environments and outperforming existing methods by 16%.
GenClaw: Code-Driven Agentic Image Generation
GenClaw introduces a code-driven agentic image generation framework that breaks the black-box paradigm by mimicking the human creative process: conceptualizing, sketching with code (SVG/HTML/Three.js), and then using generative models for texture and photorealism.
EVA01: Unified Native 3D Understanding and Generation via Mixture-of-Transformers
EVA01 is a unified framework that integrates 3D mesh as a native modality into multimodal language models via a Mixture-of-Transformers architecture, enabling state-of-the-art text-to-3D generation and long-context multi-turn geometric editing.
FFAvatar: Few-Shot, Feed-Forward, and Generalizable Avatar Reconstruction
FFAvatar proposes a feed-forward framework for reconstructing high-quality, animatable 3D Gaussian head avatars from few unposed images in seconds, achieving a 5.5 PSNR improvement over state-of-the-art on the NeRSemble benchmark.
NeuROK: Generative 4D Neural Object Kinematics
This paper introduces NeuROK, a data-driven approach for generative 4D neural object kinematics that learns a latent space and transformer-based encoder-decoder to simulate realistic temporal deformations of static objects under various physical conditions, overcoming limitations of predefined physical models.