mlsanityData to 14 Oct 2026

VisHall3D: Monocular Semantic Scene Completion from Reconstructing the Visible Regions to Hallucinating the Invisible Regions

arXiv 2507.19188 ↗First posted 25 Jul 2025cs.CVcited by 7

Not on the map. Why?

Abstract (arXiv)

This paper introduces VisHall3D, a novel two-stage framework for monocular semantic scene completion that aims to address the issues of feature entanglement and geometric inconsistency prevalent in existing methods. VisHall3D decomposes the scene completion task into two stages: reconstructing the visible regions (vision) and inferring the invisible regions (hallucination). In the first stage, VisFrontierNet, a visibility-aware projection module, is introduced to accurately trace the visual frontier while preserving fine-grained details. In the second stage, OcclusionMAE, a hallucination network, is employed to generate plausible geometries for the invisible regions using a noise injection mechanism. By decoupling scene completion into these two distinct stages, VisHall3D effectively mitigates feature entanglement and geometric inconsistency, leading to significantly improved reconstruction quality. The effectiveness of VisHall3D is validated through extensive experiments on two challenging benchmarks: SemanticKITTI and SSCBench-KITTI-360. VisHall3D achieves state-of-the-art performance, outperforming previous methods by a significant margin and paves the way for more accurate and reliable scene understanding in autonomous driving and other applications.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 44 of its 49 references that are in the map. First posted on arXiv 25 Jul 2025.

Builds on

44 of 49 references.

References · 44

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020End-to-End Object Detection with Transformerscited by 7,414
  3. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  4. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
  5. 2015Fast R-CNNcited by 2,933
  6. 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 2,010
  7. 2022BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformerscited by 1,218
  8. 2019Objects as Pointscited by 1,175
  9. 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,044
  10. 2021KITTI-360: A Novel Dataset and Benchmarks for Urban Scene Understanding in 2D and 3Dcited by 575
  11. 2020Sparse R-CNN: End-to-End Object Detection with Learnable Proposalscited by 495
  12. 2016Semantic Scene Completion from a Single Depth Imagecited by 472
32 more
  1. 2019Cascade R-CNN: High Quality Object Detection and Instance Segmentationcited by 429
  2. 2023Tri-Perspective View for Vision-Based 3D Semantic Occupancy Predictioncited by 409
  3. 2021MonoScene: Monocular 3D Semantic Scene Completioncited by 363
  4. 2022Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentationcited by 344
  5. 2023SurroundOcc: Multi-Camera 3D Occupancy Prediction for Autonomous Drivingcited by 316
  6. 2023VoxFormer: Sparse Voxel Transformer for Camera-based 3D Semantic Scene Completioncited by 313
  7. 2023OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy Predictioncited by 293
  8. 2020Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completioncited by 223
  9. 2020LMSCNet: Lightweight Multiscale 3D Semantic Completioncited by 188
  10. 2021Objects are Different: Flexible Monocular 3D Object Detectioncited by 185
  11. 2023FB-OCC: 3D Occupancy Prediction based on Forward-Backward View Transformationcited by 165
  12. 2020Anisotropic Convolutional Networks for 3D Semantic Scene Completioncited by 127
  13. 2020S3CNet: A Sparse Semantic Scene Completion Network for LiDAR Point Cloudscited by 102
  14. 2023Symphonize 3D Semantic Scene Completion with Contextual Instance Queriescited by 90
  15. 2023NDC-Scene: Boost Monocular 3D Semantic Scene Completion in Normalized Device Coordinates Spacecited by 90
  16. 2021AutoShape: Real-Time Shape-Aware Monocular 3D Object Detectioncited by 87
  17. 2022MonoDETR: Depth-guided Transformer for Monocular 3D Object Detectioncited by 86
  18. 2023OccDepth: A Depth-Aware Method for 3D Semantic Scene Completioncited by 86
  19. 2021MobileStereoNet: Towards Lightweight Deep Networks for Stereo Matchingcited by 82
  20. 2023EPro-PnP: Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimationcited by 79
  21. 2023SSCBench: A Large-Scale 3D Semantic Scene Completion Benchmark for Autonomous Drivingcited by 75
  22. 2020Semantic Scene Completion using Local Deep Implicit Functions on LiDAR Datacited by 74
  23. 2024Context and Geometry Aware Voxel Transformer for Semantic Scene Completioncited by 50
  24. 2022Diversity Matters: Fully Exploiting Depth Clues for Reliable Monocular 3D Object Detectioncited by 46
  25. 2024MonoOcc: Digging into Monocular Semantic Occupancy Predictioncited by 44
  26. 2024Not All Voxels Are Equal: Hardness-Aware Semantic Scene Completion with Self-Distillationcited by 43
  27. 2021M3DSSD: Monocular 3D Single Stage Object Detectorcited by 39
  28. 2020Depth Based Semantic Scene Completion with Position Importance Aware Losscited by 37
  29. 2024Hierarchical Temporal Context Learning for Camera-based Semantic Scene Completioncited by 32
  30. 2023Bridging Stereo Geometry and BEV Representation with Reliable Mutual Interaction for Semantic Scene Completioncited by 22
  31. 2024SPIdepth: Strengthened Pose Information for Self-supervised Monocular Depth Estimationcited by 8
  32. 2024OccTransformer: Improving BEVFormer for 3D camera-only occupancy predictioncited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 5.

7 papers.

  1. 2025OneOcc: Semantic Occupancy Prediction for Legged Robots with a Single Panoramic Cameracited by 8
  2. 2025VOIC: Visible-Occluded Integrated Guidance for 3D Semantic Scene Completioncited by 0
  3. 2026Adaptive Multi-Scale Channel-Spatial Attention Aggregation Framework for 3D Indoor Semantic Scene Completion Toward Assisting Visually Impairedcited by 0
  4. 2026Sparsity-Aware Voxel Attention and Foreground Modulation for 3D Semantic Scene Completioncited by 0
  5. 2026Semantic Occupancy Prediction with Dual Range-Voxel Representationcited by 0
  6. 2026Geometry-Aware Spatio-Temporal Context Modeling for 4D Occupancy Forecastingcited by 0
  7. 2026SSC-Priors: Exploring Semantic and Visibility Priors to Boost Lidar Semantic Scene Completioncited by 0