mlsanityData to 7 Oct 2026

Adaptive Multi-Scale Channel-Spatial Attention Aggregation Framework for 3D Indoor Semantic Scene Completion Toward Assisting Visually Impaired

arXiv 2602.16385 ↗First posted 18 Feb 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Independent indoor mobility remains a critical challenge for individuals with visual impairments, largely due to the limited capability of existing assistive systems in detecting fine-grained hazardous objects such as chairs, tables, and small obstacles. These perceptual blind zones substantially increase the risk of collision in unfamiliar environments. To bridge the gap between monocular 3D vision research and practical assistive deployment, this paper proposes an Adaptive Multi-scale Attention Aggregation (AMAA) framework for monocular 3D semantic scene completion using only a wearable RGB camera. The proposed framework addresses two major limitations in 2D-to-3D feature lifting: noise diffusion during back-projection and structural instability in multi-scale fusion. A parallel channel--spatial attention mechanism is introduced to recalibrate lifted features along semantic and geometric dimensions, while a hierarchical adaptive gating strategy regulates cross-scale information flow to preserve fine-grained structural details. Experiments on the NYUv2 benchmark demonstrate that AMAA achieves an overall mIoU of 27.88%. Crucially, it yields significant relative improvements of 16.9% for small objects and 10.4% for tables over the MonoScene baseline. Furthermore, a wearable prototype based on an NVIDIA Jetson Orin NX and a ZED~2i camera validates stable real-time performance in indoor environments, demonstrating the feasibility of deploying monocular 3D scene completion for assistive navigation.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 36 of its 64 references that are in the map. First posted on arXiv 18 Feb 2026.

Builds on

36 of 64 references.

References · 36

  1. 2016Semantic Scene Completion from a Single Depth Imagecited by 468
  2. 2022LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Modelscited by 459
  3. 2023Tri-Perspective View for Vision-Based 3D Semantic Occupancy Predictioncited by 404
  4. 2021MonoScene: Monocular 3D Semantic Scene Completioncited by 357
  5. 2023Occ3D: A Large-Scale 3D Occupancy Prediction Benchmark for Autonomous Drivingcited by 325
  6. 2023VoxFormer: Sparse Voxel Transformer for Camera-based 3D Semantic Scene Completioncited by 306
  7. 2023OccWorld: Learning a 3D Occupancy World Model for Autonomous Drivingcited by 230
  8. 2023OpenOccupancy: A Large Scale Benchmark for Surrounding Semantic Occupancy Perceptioncited by 218
  9. 2020LMSCNet: Lightweight Multiscale 3D Semantic Completioncited by 187
  10. 2022Hydra: A Real-time Spatial Perception System for 3D Scene Graph Construction and Optimizationcited by 175
  11. 2024GaussianFormer: Scene as Gaussians for Vision-Based 3D Semantic Occupancy Predictioncited by 155
  12. 2023SelfOcc: Self-Supervised Vision-Based 3D Occupancy Predictioncited by 133
24 more
  1. 2023Robo3D: Towards Robust and Reliable 3D Perception against Corruptionscited by 128
  2. 2020Anisotropic Convolutional Networks for 3D Semantic Scene Completioncited by 126
  3. 20203D Sketch-aware Semantic Scene Completion via Semi-supervised Structure Priorcited by 124
  4. 2023FlashOcc: Fast and Memory-Efficient Occupancy Prediction via Channel-to-Height Plugincited by 101
  5. 20213D Semantic Scene Completion: a Surveycited by 74
  6. 2024POP-3D: Open-Vocabulary 3D Occupancy Prediction from Imagescited by 52
  7. 2024Co-Occ: Coupling Explicit Feature Fusion with Volume Rendering Regularization for Multi-Modal 3D Semantic Occupancy Predictioncited by 46
  8. 2024EmbodiedOcc: Embodied 3D Occupancy Prediction for Vision-based Online Scene Understandingcited by 43
  9. 2023OccNeRF: Advancing 3D Occupancy Prediction in LiDAR-Free Environmentscited by 39
  10. 2024RadarOcc: Robust 3D Occupancy Prediction with 4D Imaging Radarcited by 32
  11. 2024Monocular Occupancy Prediction for Scalable Indoor Scenescited by 29
  12. 2024OccMamba: Semantic Occupancy Prediction with State Space Modelscited by 29
  13. 2025EmbodiedOcc++: Boosting Embodied 3D Occupancy Prediction with Plane Regularization and Uncertainty Samplercited by 26
  14. 2024GEOcc: Geometrically Enhanced 3D Occupancy Network with Implicit-Explicit Depth Fusion and Contextual Self-Supervisioncited by 21
  15. 2024ProtoOcc: Accurate, Efficient 3D Occupancy Prediction Using Dual Branch Encoder-Prototype Query Decodercited by 17
  16. 2024DiffSSC: Semantic LiDAR Scan Completion using Denoising Diffusion Probabilistic Modelscited by 16
  17. 2025Skip Mamba Diffusion for Monocular 3D Semantic Scene Completioncited by 9
  18. 2024Hierarchical Context Alignment with Disentangled Geometric and Temporal Modeling for Semantic Occupancy Predictioncited by 8
  19. 2025SliceOcc: Indoor 3D Semantic Occupancy Prediction with Vertical Slice Representationcited by 8
  20. 2024Hardness-Aware Scene Synthesis for Semi-Supervised 3D Object Detectioncited by 7
  21. 2025VisHall3D: Monocular Semantic Scene Completion from Reconstructing the Visible Regions to Hallucinating the Invisible Regionscited by 7
  22. 2024LightOcc: Lightweight Spatial Embedding for Efficient Vision-based 3D Occupancy Predictioncited by 6
  23. 2024Unleashing Network Potentials for Semantic Scene Completioncited by 4
  24. 2024How Can Haptic Feedback Assist People with Blind and Low Vision (BLV): A Systematic Literature Reviewcited by 2

Built on by

Not cited yet.