Research map / Detection and segmentation / 3D detection and driving
(AF)2-S3Net: Attentive Feature Fusion With Adaptive Feature Selection for Sparse Semantic Segmentation Network
Ran Cheng, Ryan Razani, Ehsan Taghavi, Enxu Li, Bingbing Liu
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add AF2M to the MinkNet42 baseline (ablation) | MinkNet42 · SemanticKITTI validation seq 08 | mIoU: 59.8 -> 65.1; increase of 5.3% | Improved |
| Add AFSM using AF2M branch outputs (ablation) | MinkNet42 baseline · SemanticKITTI validation seq 08 | mIoU: 59.8 -> 63.3; increase of 3.5% | Improved |
| Add AF2M and AFSM together (ablation) | MinkNet42 · SemanticKITTI validation seq 08 | mIoU: 59.8 -> 68.6; increase of 8.8% | Improved |
| Add Lovász loss (ablation) | AF2M + AFSM · SemanticKITTI validation seq 08 | mIoU: 68.6 -> 70.2 | Improved |
| Add geo-aware anisotropic loss with Lovász loss (ablation) | AF2M + AFSM · SemanticKITTI validation seq 08 | mIoU: 70.2 -> 74.2 | Improved |
| Use (AF)²-S3Net instead of SPVNAS | SemanticKITTI test · SemanticKITTI test | mIoU: 67.0 -> 69.7 | Improved |
| Use (AF)²-S3Net instead of MinkNet42 | nuScenes-lidarseg · nuScenes-lidarseg validation | Mean IoU: 60.8 -> 62.2 | Improved |
Closest papers by meaning
- GP-S3Net: Graph-Based Panoptic Sparse Semantic Segmentation Network (ICCV 2021 · similarity 0.90)
- SVQNet: Sparse Voxel-Adjacent Query Network for 4D Spatio-Temporal LiDAR Semantic Segmentation (ICCV 2023 · similarity 0.89)
- Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-Based Perception (TPAMI 2021 · similarity 0.89)
- Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation (CVPR 2021 · similarity 0.89)
- Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Data (CVPR 2022 · similarity 0.89)
- PC-BEV: An Efficient Polar-Cartesian BEV Fusion Framework for LiDAR Semantic Segmentation (AAAI 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
