Thinking in Blender: Staged Executable Inverse Graphics with Vision-Language Models

• URL: http://arxiv.org/abs/2606.02580v1
• Date: Mon, 01 Jun 2026 17:59:53 GMT
• Title: Thinking in Blender: Staged Executable Inverse Graphics with Vision-Language Models
• Abstract summary: We introduce Staged Executable Inverse Graphics (SEIG), an agentic framework that reconstructs a 3D scene from a single image.<n>We evaluate our framework across diverse scenes using a range of reconstruction metrics spanning pixel-level, perceptual, and semantic fidelity.
• Score: 29.14307229425815
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Inverse graphics is a longstanding and highly underconstrained problem that seeks to reconstruct images as editable 3D scenes which can be rendered, relit, and manipulated. In this work, we investigate whether pretrained vision-language models (VLMs) can perform executable inverse graphics directly from a single image by reconstructing a scene as an editable Blender program, without relying on specialized 2D or 3D foundation models, differentiable rendering, or multi-view supervision. We introduce Staged Executable Inverse Graphics (SEIG), an agentic framework that reconstructs a 3D scene from a single image by progressively refining scene factors including geometry, materials, composition, and lighting directly in executable Blender code space. We evaluate our framework across diverse scenes using a range of reconstruction metrics spanning pixel-level, perceptual, and semantic fidelity. Our experiments show that staged reconstruction substantially improves reconstruction fidelity, highlighting the importance of task decomposition for executable inverse graphics with general-purpose VLMs. Finally, we showcase various downstream applications enabled by the reconstructed editable Blender scenes.

Related papers

• SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation [50.175987640001644]
  We present SimWorlds: a multi-agent framework that produces dynamic, editable 4D scenes from text.<n>We also introduce 4DBuildBench, a benchmark for assessing both visual fidelity and physical consistency of the procedural dynamic 3D scenes.
  arXiv  Detail & Related papers  (2026-07-02T06:29:18Z)
• SceneExpander: Expanding 3D Scenes with Free-Form Inserted Views [69.08965991211704]
  We study 3D scene expansion in a user-centric workflow, starting from a real scene captured by multi-view images.<n>Unlike simple object editing or style transfer in a fixed scene, the inserted view is often 3D-misaligned with the original reconstruction.<n>We propose SceneExpander, which applies test-time adaptation to a parametric feed-forward 3D reconstruction model.
  arXiv  Detail & Related papers  (2026-03-28T02:04:48Z)
• Vision-as-Inverse-Graphics Agent via Interleaved Multimodal Reasoning [105.35082963701541]
  VIGA (Vision-as-Inverse-Graphic Agent) reconstructs or edits scenes through a closed-loop write-run-render-compare-revise procedure.<n>To support long-horizon reasoning, VIGA combines (i) a skill library that alternates generator and verifier roles and (ii) an evolving context memory.
  arXiv  Detail & Related papers  (2026-01-16T09:11:55Z)
• 3D-RE-GEN: 3D Reconstruction of Indoor Scenes with a Generative Framework [7.4570191712029965]
  3D-RE-GEN is a compositional framework that reconstructs a single image into textured 3D objects and a background.<n>Our pipeline integrates models for asset detection, reconstruction, and placement, pushing certain models beyond their originally intended domains.<n>Unlike current methods, 3D-RE-GEN generates a comprehensive background that spatially constrains objects during optimization.
  arXiv  Detail & Related papers  (2025-12-19T11:20:52Z)
• Constructing a 3D Scene from a Single Image [31.11317559252235]
  SceneFuse-3D is a training-free framework designed to synthesize coherent 3D scenes from a single top-down view.<n>We decompose the input image into overlapping regions and generate each using a pretrained 3D object generator.<n>This modular design allows us to overcome resolution bottlenecks and preserve spatial structure without requiring 3D supervision or fine-tuning.
  arXiv  Detail & Related papers  (2025-05-21T17:10:47Z)
• BlenderAlchemy: Editing 3D Graphics with Vision-Language Models [4.852796482609347]
  A vision-based edit generator and state evaluator work together to find the correct sequence of actions to achieve the goal. Inspired by the role of visual imagination in the human design process, we supplement the visual reasoning capabilities of Vision-Language Models with "imagined" reference images.
  arXiv  Detail & Related papers  (2024-04-26T19:37:13Z)
• Re-Thinking Inverse Graphics With Large Language Models [51.333105116400205]
  Inverse graphics -- inverting an image into physical variables that, when rendered, enable reproduction of the observed scene -- is a fundamental challenge in computer vision and graphics. We propose the Inverse-Graphics Large Language Model (IG-LLM), an inversegraphics framework centered around an LLM. We incorporate a frozen pre-trained visual encoder and a continuous numeric head to enable end-to-end training.
  arXiv  Detail & Related papers  (2024-04-23T16:59:02Z)
• Single-view 3D Scene Reconstruction with High-fidelity Shape and Texture [47.44029968307207]
  We propose a novel framework for simultaneous high-fidelity recovery of object shapes and textures from single-view images. Our approach utilizes the proposed Single-view neural implicit Shape and Radiance field (SSR) representations to leverage both explicit 3D shape supervision and volume rendering. A distinctive feature of our framework is its ability to generate fine-grained textured meshes while seamlessly integrating rendering capabilities into the single-view 3D reconstruction model.
  arXiv  Detail & Related papers  (2023-11-01T11:46:15Z)
• Differentiable Blocks World: Qualitative 3D Decomposition by Rendering Primitives [70.32817882783608]
  We present an approach that produces a simple, compact, and actionable 3D world representation by means of 3D primitives. Unlike existing primitive decomposition methods that rely on 3D input data, our approach operates directly on images. We show that the resulting textured primitives faithfully reconstruct the input images and accurately model the visible 3D points.
  arXiv  Detail & Related papers  (2023-07-11T17:58:31Z)
• High-fidelity 3D GAN Inversion by Pseudo-multi-view Optimization [51.878078860524795]
  We present a high-fidelity 3D generative adversarial network (GAN) inversion framework that can synthesize photo-realistic novel views. Our approach enables high-fidelity 3D rendering from a single image, which is promising for various applications of AI-generated 3D content.
  arXiv  Detail & Related papers  (2022-11-28T18:59:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.