mlsanityData to 14 Oct 2026

CGGS: Consistency-Augmented Geometric Gaussian Splatting for Ego-Centric 3D Scene Generation

arXiv 2607.03819 ↗First posted 4 Jul 2026cs.GR cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Challenges remain in ego-centric 3D scene generation due to limited view overlap and the dominant influence of individual perspectives on scene interpretation. These factors hinder the creation of viewpoint-consistent and semantically aligned visual content, as well as the construction of accurate geometric structures. In this paper, we propose CGGS, a text-to-3D framework aiming to enhance 3D-content-awareness and address geometric distortions in ego-centric scene generation. Firstly, the Ego-centric Generator is proposed by fine-tuning a Multi-View Latent Diffusion Model with consistency-augmented loss to generate consistent, high-fidelity 2D content aligned with textual descriptions. Then, Layout Decorator leverages optical flow and point-track correspondence to estimate depth, therefore producing dense point clouds as coarse layouts from the ego-centric 2D priors. Building on this initialization, Geometric Refiner is proposed to enhance 3D Gaussian reconstruction via an entropy-based Mutual Information Depth Loss (MID) combined with a hierarchical optimization scheme for improving visual quality and geometric structure. Comprehensive experiments demonstrate that CGGS outperforms previous methods in generating coherent and accurate text-driven 3D scenes.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 76 of its 91 references that are in the map. First posted on arXiv 4 Jul 2026.

Builds on

76 of 91 references.

References · 76

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2023GPT-4 Technical Reportcited by 18,396
  5. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  6. 2020Denoising Diffusion Implicit Modelscited by 9,781
  7. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  8. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  9. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  10. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  11. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  12. 2022Classifier-Free Diffusion Guidancecited by 5,847
38 more
  1. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
  2. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,101
  3. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,614
  4. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,611
  5. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  6. 2022DreamFusion: Text-to-3D using 2D Diffusioncited by 3,025
  7. 2021CLIPScore: A Reference-free Evaluation Metric for Image Captioningcited by 2,424
  8. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  9. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  10. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,584
  11. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,495
  12. 2017Matterport3D: Learning from RGB-D Data in Indoor Environmentscited by 1,376
  13. 2024Depth Anything: Unleashing the Power of Large-Scale Unlabeled Datacited by 1,360
  14. 2022Magic3D: High-Resolution Text-to-3D Content Creationcited by 1,290
  15. 2023ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillationcited by 1,219
  16. 2022Exploring CLIP for Assessing the Look and Feel of Imagescited by 991
  17. 2023MVDream: Multi-view Diffusion for 3D Generationcited by 938
  18. 2024LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creationcited by 739
  19. 2023Fantasia3D: Disentangling Geometry and Appearance for High-quality Text-to-3D Content Creationcited by 687
  20. 2023pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstructioncited by 684
  21. 2023ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depthcited by 657
  22. 2021Common Objects in 3D: Large-Scale Learning and Evaluation of Real-life 3D Category Reconstructioncited by 652
  23. 2022Score Jacobian Chaining: Lifting Pretrained 2D Diffusion Models for 3D Generationcited by 597
  24. 2024MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Imagescited by 566
  25. 2016Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modelingcited by 558
  26. 2023MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generationcited by 536
  27. 2023CoTracker: It is Better to Track Togethercited by 522
  28. 2021Zero-Shot Text-Guided Object Generation with Dream Fieldscited by 513
  29. 2022Latent-NeRF for Shape-Guided Generation of 3D Shapes and Texturescited by 506
  30. 2023Q-Align: Teaching LMMs for Visual Scoring via Discrete Text-Defined Levelscited by 503
  31. 2023Splatter Image: Ultra-Fast Single-View 3D Reconstructioncited by 364
  32. 2024GS-LRM: Large Reconstruction Model for 3D Gaussian Splattingcited by 349
  33. 2021Text2Mesh: Text-Driven Neural Stylization for Meshescited by 334
  34. 2024GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generationcited by 301
  35. 2023FSGS: Real-Time Few-shot View Synthesis using Gaussian Splattingcited by 279
  36. 2024DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalizationcited by 256
  37. 2023Text2Room: Extracting Textured 3D Meshes from 2D Text-to-Image Modelscited by 245
  38. 2022NoPe-NeRF: Optimising Neural Radiance Field with No Pose Priorcited by 235

50 most cited of 76 All 76 (JSON)

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025G4Splat: Geometry-Guided Gaussian Splatting with Generative Prior
  2. 2025Layout2Scene: 3D Semantic Layout Guided Scene Generation via Geometry and Appearance Diffusion Priors
  3. 2025BloomScene: Lightweight Structured 3D Gaussian Splatting for Crossmodal Scene Generation
  4. 2026Scene Generation at Absolute Scale: Utilizing Semantic and Geometric Guidance From Text for Accurate and Interpretable 3D Indoor Scene Generation
  5. 2026PSGS: Text-driven Panorama Sliding Scene Generation via Gaussian Splatting
  6. 2025Bridging Geometry-Coherent Text-to-3D Generation with Multi-View Diffusion Priors and Gaussian Splatting
  7. 2024GaussCtrl: Multi-View Consistent Text-Driven 3D Gaussian Splatting Editing
  8. 2025Geometry and Perception Guided Gaussians for Multiview-consistent 3D Generation from a Single Image