mlsanityData to 14 Oct 2026

Total-Decom: Decomposed 3D Scene Reconstruction with Minimal Interaction

arXiv 2403.19314 ↗First posted 28 Mar 2024cs.CVcited by 12

Not on the map. Why?

Abstract (arXiv)

Scene reconstruction from multi-view images is a fundamental problem in computer vision and graphics. Recent neural implicit surface reconstruction methods have achieved high-quality results; however, editing and manipulating the 3D geometry of reconstructed scenes remains challenging due to the absence of naturally decomposed object entities and complex object/background compositions. In this paper, we present Total-Decom, a novel method for decomposed 3D reconstruction with minimal human interaction. Our approach seamlessly integrates the Segment Anything Model (SAM) with hybrid implicit-explicit neural surface representations and a mesh-based region-growing technique for accurate 3D object decomposition. Total-Decom requires minimal human annotations while providing users with real-time control over the granularity and quality of decomposition. We extensively evaluate our method on benchmark datasets and demonstrate its potential for downstream applications, such as animation and scene editing. The code is available at https://github.com/CVMI-Lab/Total-Decom.git.

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Builds on 39 of its 48 references that are in the map. First posted on arXiv 28 Mar 2024.

Builds on

39 of 48 references.

References · 39

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  3. 2023Segment Anythingcited by 8,234
  4. 2017Mask R-CNNcited by 7,277
  5. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  6. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
  7. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  8. 2021Masked-attention Mask Transformer for Universal Image Segmentationcited by 2,351
  9. 2021NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstructioncited by 1,712
  10. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,584
  11. 2021Volume Rendering of Neural Implicit Surfacescited by 996
  12. 2019The Replica Dataset: A Digital Replica of Indoor Spacescited by 904
27 more
  1. 2020Convolutional Occupancy Networkscited by 761
  2. 2022Language-driven Semantic Segmentationcited by 717
  3. 2020Implicit Geometric Regularization for Learning Shapescited by 710
  4. 2021UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstructioncited by 646
  5. 2021NICE-SLAM: Neural Implicit Scalable Encoding for SLAMcited by 589
  6. 2023LERF: Language Embedded Radiance Fieldscited by 543
  7. 2022MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstructioncited by 470
  8. 2021In-Place Scene Labelling and Understanding with Implicit Scene Representationcited by 415
  9. 2020Local Implicit Grid Representations for 3D Scenescited by 373
  10. 2022Decomposing NeRF for Editing via Feature Field Distillationcited by 371
  11. 2021Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scanscited by 336
  12. 2022Panoptic Neural Fields: A Semantic Object-Aware Neural Scene Representationcited by 243
  13. 2023Text2Tex: Text-driven Texture Synthesis via Diffusion Modelscited by 235
  14. 2022Neural Feature Fusion Fields: 3D Distillation of Self-Supervised 2D Image Representationscited by 196
  15. 2022Panoptic Lifting for 3D Scene Understanding with Neural Fieldscited by 174
  16. 2022NeuRIS: Neural Reconstruction of Indoor Scenes Using Normal Priorscited by 125
  17. 2022Neural 3D Scene Reconstruction with the Manhattan-world Assumptioncited by 123
  18. 2023Neural Kernel Surface Reconstructioncited by 104
  19. 2022Object-Compositional Neural Implicit Surfacescited by 74
  20. 2022DM-NeRF: 3D Scene Geometry Decomposition and Manipulation from 2D Imagescited by 69
  21. 2023vMAP: Vectorised Object Mapping for Neural Field SLAMcited by 67
  22. 2022Interactive Segmentation of Radiance Fieldscited by 60
  23. 2023ObjectSDF++: Improved Object-Compositional Neural Implicit Surfacescited by 39
  24. 2023RICO: Regularizing the Unobservable for Indoor Compositional Reconstructioncited by 20
  25. 2023MarS3D: A Plug-and-Play Motion-Aware Model for Semantic Segmentation on Multi-Scan 3D Point Cloudscited by 19
  26. 2023Learning a Room with the Occ-SDF Hybrid: Signed Distance Function Mingled with Occupancy Aids Scene Representationcited by 10
  27. 2023AssetField: Assets Mining and Reconfiguration in Ground Feature Plane Representationcited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 7; 2026: 2.

12 papers.

  1. 2025Decompositional Neural Scene Reconstruction with Generative Diffusion Priorcited by 31
  2. 2025LiteReality: Graphics-Ready 3D Scene Reconstruction from RGB-D Scanscited by 21
  3. 2024Deformable Radial Kernel Splattingcited by 17
  4. 2024How Far are AI-generated Videos from Simulating the 3D Visual World: A Learned 3D Evaluation Approachcited by 16
  5. 2025Trace3D: Consistent Segmentation Lifting via Gaussian Instance Tracingcited by 13
  6. 2024SAMa: Material-aware 3D Selection and Segmentationcited by 10
  7. 2025Aligning Effective Tokens with Video Anomaly in Large Language Modelscited by 8
  8. 2025Equipping Vision Foundation Model with Mixture of Experts for Out-of-Distribution Detectioncited by 5
  9. 2025Physics-Based 3D Simulation for Synthetic Data Generation and Failure Analysis in Packaging Stability Assessmentcited by 0
  10. 2025Proactive Scene Decomposition and Reconstructioncited by 0
  11. 2026Learning to See through Illumination Extremes with Event Streaming in Multimodal Large Language Modelscited by 0
  12. 2026First Shape, Then Meaning: Efficient Geometry and Semantics Learning for Indoor Reconstructioncited by 0