mlsanityData to 7 Oct 2026

Fore-Mamba3D: Mamba-based Foreground-Enhanced Encoding for 3D Object Detection

arXiv 2602.19536 ↗First posted 23 Feb 2026cs.CV cs.AIcited by 0

On the map: State-space sequence models (S4, Mamba) → Point clouds, detection and remote sensing

Abstract (arXiv)

Linear modeling methods like Mamba have been merged as the effective backbone for the 3D object detection task. However, previous Mamba-based methods utilize the bidirectional encoding for the whole non-empty voxel sequence, which contains abundant useless background information in the scenes. Though directly encoding foreground voxels appears to be a plausible solution, it tends to degrade detection performance. We attribute this to the response attenuation and restricted context representation in the linear modeling for fore-only sequences. To address this problem, we propose a novel backbone, termed Fore-Mamba3D, to focus on the foreground enhancement by modifying Mamba-based encoder. The foreground voxels are first sampled according to the predicted scores. Considering the response attenuation existing in the interaction of foreground voxels across different instances, we design a regional-to-global slide window (RGSW) to propagate the information from regional split to the entire sequence. Furthermore, a semantic-assisted and state spatial fusion module (SASFMamba) is proposed to enrich contextual representation by enhancing semantic and geometric awareness within the Mamba model. Our method emphasizes foreground-only encoding and alleviates the distance-based and causal dependencies in the linear autoregression model. The superior performance across various benchmarks demonstrates the effectiveness of Fore-Mamba3D in the 3D object detection task.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Point clouds, detection and remote sensing, one of 10 directions in State-space sequence models (S4, Mamba). Builds on 41 of its 53 references that are in the map. First posted on arXiv 23 Feb 2026.

Builds on

41 of 53 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024LION: Linear Group RNN for 3D Object Detection in Point Clouds25 Jul
  2. 2024Voxel Mamba: Group-Free State Space Models for Point Cloud based 3D Object Detection15 Jun
  3. 2024PointMamba: A Simple State Space Model for Point Cloud Analysis16 Feb

Other references · 38

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  3. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  4. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  5. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  6. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,436
  7. 2017VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detectioncited by 1,323
  8. 2020Center-based 3D Object Detection and Trackingcited by 1,143
  9. 2024Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Modelcited by 1,051
  10. 2024VMamba: Visual State Space Modelcited by 955
  11. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 939
  12. 2019PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detectioncited by 890
26 more
  1. 2023RWKV: Reinventing RNNs for the Transformer Eracited by 814
  2. 2021DETR3D: 3D Object Detection from Multi-view Images via 3D-to-2D Queriescited by 655
  3. 2023Retentive Network: A Successor to Transformer for Large Language Modelscited by 594
  4. 2022TransFusion: Robust LiDAR-Camera Fusion for 3D Object Detection with Transformerscited by 464
  5. 2019PointPainting: Sequential Fusion for 3D Object Detectioncited by 449
  6. 2020Voxel R-CNN: Towards High Performance Voxel-based 3D Object Detectioncited by 409
  7. 2019A Survey of Deep Learning Techniques for Autonomous Drivingcited by 401
  8. 2019From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Networkcited by 366
  9. 2023VoxelNeXt: Fully Sparse VoxelNet for 3D Object Detection and Trackingcited by 221
  10. 2021Embodied Intelligence via Learning and Evolutioncited by 154
  11. 2022Focal Sparse Convolutional Networks for 3D Object Detectioncited by 142
  12. 20223D Object Detection for Autonomous Driving: A Comprehensive Surveycited by 132
  13. 2022Not All Points Are Equal: Learning Highly Efficient Point-based Detectors for 3D LiDAR Point Cloudscited by 131
  14. 2023DSVT: Dynamic Sparse Voxel Transformer with Rotated Setscited by 122
  15. 2019TANet: Robust 3D Object Detection from Point Clouds with Triple Attentioncited by 108
  16. 2021Improving 3D Object Detection with Channel-wise Transformercited by 108
  17. 2022Multi-modal Sensor Fusion for Auto Driving Perception: A Surveycited by 73
  18. 2024SAFDNet: A Simple and Effective Network for Fully Sparse 3D Object Detectioncited by 52
  19. 2023HEDNet: A Hierarchical Encoder-Decoder Network for 3D Object Detection in Point Cloudscited by 45
  20. 2022RBGNet: Ray-based Grouping for 3D Object Detectioncited by 40
  21. 2021From Voxel to Point: IoU-guided 3D Object Detection for Point Cloud with Voxel-to-Point Decodercited by 23
  22. 2023Onboard dynamic-object detection and tracking for autonomous robot navigation with RGB-D cameracited by 13
  23. 2024ScatterFormer: Efficient Voxel Transformer with Scattered Linear Attentioncited by 11
  24. 2025FSHNet: Fully Sparse Hybrid Network for 3D Object Detectioncited by 9
  25. 2024SeSame: Simple, Easy 3D Object Detection with Point-Wise Semanticscited by 3
  26. 2024MambaDETR: Query-based Temporal Modeling using State Space Model for Multi-View 3D Object Detectioncited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025WinMamba: Multi-Scale Shifted Windows in State Space Model for 3D Object Detection
  2. 2025MambaFusion: Height-Fidelity Dense Global Fusion for Multi-modal 3D Object Detection
  3. 2025MambaNeXt-YOLO: A Hybrid State Space Model for Real-time Object Detection
  4. 2024MambaBEV: An EV-based 3D detection model with Mamba2
  5. 2025DA-Mamba: Domain Adaptive Hybrid Mamba-Transformer Based One-Stage Object Detection
  6. 2026MambaPSA: A Mamba-based Replacement for C2PSA in YOLO26
  7. 2026SF-Mamba: Rethinking State Space Model for Vision
  8. 2025MambaEye: A Size-Agnostic Visual Encoder with Causal Sequential Processing