BlendFusion -- Scalable Synthetic Data Generation for Diffusion Model Training

• URL: http://arxiv.org/abs/2604.09022v1
• Date: Fri, 10 Apr 2026 06:36:38 GMT
• Title: BlendFusion -- Scalable Synthetic Data Generation for Diffusion Model Training
• Abstract summary: We propose BlendFusion, a framework for synthetic data generation from 3D scenes using path tracing.<n>Our pipeline incorporates an object-centric camera placement strategy, robust filtering mechanisms, and automatic captioning.<n>We empirically analyze the quality of FineBLEND, an image-caption dataset constructed from a diverse set of 3D scenes.
• Score: 0.07419401583272026
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: With the rapid adoption of diffusion models, synthetic data generation has emerged as a promising approach for addressing the growing demand for large-scale image datasets. However, images generated purely by diffusion models often exhibit visual inconsistencies, and training models on such data can create an autophagous feedback loop that leads to model collapse, commonly referred to as Model Autophagy Disorder (MAD). To address these challenges, we propose BlendFusion, a scalable framework for synthetic data generation from 3D scenes using path tracing. Our pipeline incorporates an object-centric camera placement strategy, robust filtering mechanisms, and automatic captioning to produce high-quality image-caption pairs. Using this pipeline, we curate FineBLEND, an image-caption dataset constructed from a diverse set of 3D scenes. We empirically analyze the quality of FineBLEND and compare it to several widely used image-caption datasets. We also demonstrate the effectiveness of our object-centric camera placement strategy relative to object-agnostic sampling approaches. Our open-source framework is designed for high configurability, enabling the community to create their own datasets from 3D scenes.

Related papers

• Sparse-to-Complete: From Sparse Image Captures to Complete 3D Scenes [32.47917212506622]
  S2C-3D is a novel sparse-view 3D reconstruction framework for high-fidelity and complete scene reconstruction from as few as six to eight images.<n>Our framework features three components: a specialized diffusion model for scene-specific image restoration, a training-free view-consistency conditioned sampling process, and a camera trajectory planning scheme.
  arXiv  Detail & Related papers  (2026-05-07T04:41:30Z)
• PoseDreamer: Scalable and Photorealistic Human Data Generation Pipeline with Diffusion Models [41.23359540842301]
  PoseDreamer is a novel pipeline that generates synthetic datasets with 3D mesh annotations.<n>We generate more than 500,000 high-quality synthetic samples, achieving a 76% improvement in image-quality metrics.
  arXiv  Detail & Related papers  (2026-03-30T17:59:18Z)
• DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion [53.70278210626701]
  We propose a data-driven multi-view reasoning approach that directly infers 3D scene geometry and camera poses from multi-view images.<n>Our framework, DiffusionSfM, parameterizes scene geometry and cameras as pixel-wise ray origins and endpoints in a global frame.<n>We empirically validate DiffusionSfM on both synthetic and real datasets, demonstrating that it outperforms classical and learning-based approaches.
  arXiv  Detail & Related papers  (2025-05-08T17:59:47Z)
• Cut-and-Splat: Leveraging Gaussian Splatting for Synthetic Data Generation [0.7864304771129751]
  We develop a synthetic data pipeline for generating context-aware instance segmentation training data for specific objects.<n>We train a Gaussian Splatting model of the target object and automatically extract the object from the video.<n>We then render the object on a random background image, and monocular depth estimation is employed to place the object in a believable pose.
  arXiv  Detail & Related papers  (2025-04-11T12:04:49Z)
• Unpaired Deblurring via Decoupled Diffusion Model [55.21345354747609]
  We propose UID-Diff, a generative-diffusion-based model designed to enhance deblurring performance on unknown domains.<n>We employ two Q-Formers as structural features and blur patterns extractors separately. The features extracted will be used for the supervised deblurring task on synthetic data and the unsupervised blur-transfer task.<n>Experiments on real-world datasets demonstrate that UID-Diff outperforms existing state-of-the-art methods in blur removal and structural preservation.
  arXiv  Detail & Related papers  (2025-02-03T17:00:40Z)
• GS-Blur: A 3D Scene-Based Dataset for Realistic Image Deblurring [50.72230109855628]
  We propose GS-Blur, a dataset of synthesized realistic blurry images created using a novel approach. We first reconstruct 3D scenes from multi-view images using 3D Gaussian Splatting (3DGS), then render blurry images by moving the camera view along the randomly generated motion trajectories. By adopting various camera trajectories in reconstructing our GS-Blur, our dataset contains realistic and diverse types of blur, offering a large-scale dataset that generalizes well to real-world blur.
  arXiv  Detail & Related papers  (2024-10-31T06:17:16Z)
• Bootstrap3D: Improving Multi-view Diffusion Model with Synthetic Data [80.92268916571712]
  A critical bottleneck is the scarcity of high-quality 3D objects with detailed captions. We propose Bootstrap3D, a novel framework that automatically generates an arbitrary quantity of multi-view images. We have generated 1 million high-quality synthetic multi-view images with dense descriptive captions.
  arXiv  Detail & Related papers  (2024-05-31T17:59:56Z)
• Diffusion Models are Efficient Data Generators for Human Mesh Recovery [55.37787289869703]
  We show that synthetic data created by generative models is complementary to CG-rendered data.<n>We propose an effective data generation pipeline based on recent diffusion models, termed HumanWild.<n>Our work could pave the way for scaling up 3D human recovery to in-the-wild scenes.
  arXiv  Detail & Related papers  (2024-03-17T06:31:16Z)
• Pyramid Diffusion for Fine 3D Large Scene Generation [56.00726092690535]
  Diffusion models have shown remarkable results in generating 2D images and small-scale 3D objects. Their application to the synthesis of large-scale 3D scenes has been rarely explored. We introduce a framework, the Pyramid Discrete Diffusion model (PDD), which employs scale-varied diffusion models to progressively generate high-quality outdoor scenes.
  arXiv  Detail & Related papers  (2023-11-20T11:24:21Z)
• Light Field Diffusion for Single-View Novel View Synthesis [32.59286750410843]
  Single-view novel view synthesis (NVS) is important but challenging in computer vision. Recent advancements in NVS have leveraged Denoising Diffusion Probabilistic Models (DDPMs) for their exceptional ability to produce high-fidelity images. We present Light Field Diffusion (LFD), a novel conditional diffusion-based approach that transcends the conventional reliance on camera pose matrices.
  arXiv  Detail & Related papers  (2023-09-20T03:27:06Z)
• IT3D: Improved Text-to-3D Generation with Explicit View Synthesis [71.68595192524843]
  This study presents a novel strategy that leverages explicitly synthesized multi-view images to address these issues. Our approach involves the utilization of image-to-image pipelines, empowered by LDMs, to generate posed high-quality images. For the incorporated discriminator, the synthesized multi-view images are considered real data, while the renderings of the optimized 3D models function as fake data.
  arXiv  Detail & Related papers  (2023-08-22T14:39:17Z)

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.