Research map / Point clouds / Segmentation and detection
PARI: Policy-Driven Active Residual Intervention for Weakly Supervised Point Cloud Segmentation
Wenhao Gao, Sen Tao, Jiawei Liu, Jiwei Feng, Yongchao Xu, Junfeng Wang, Tao Jiang, jiangbo Ai, Jin Zhang, Zheng-Jun Zha
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes PARI, a policy-driven active residual intervention framework for weakly supervised 3D point cloud semantic segmentation that actively identifies and corrects erroneous predictions rather than passively propagating pseudo-labels.
Problem
Weakly supervised semantic segmentation of point clouds from extremely sparse point-level labels; existing passive pseudo-label propagation or consistency regularization methods fail to explicitly identify and correct erroneous predictions, leading to error accumulation.
Approach
Replaces passive pseudo-label propagation and consistency regularization with an active, policy-driven intervention that explicitly detects erroneous predictions and applies residual corrections. Only the abstract was available, so finer mechanism details are not stated.
Limitations
- Only the abstract was available; experimental details unknown
Closest papers by meaning
- Weakly Supervised Point Cloud Semantic Segmentation via Artificial Oracle (CVPR 2024 · similarity 0.91)
- SQN: Weakly-Supervised Semantic Segmentation of Large-Scale 3D Point Clouds (ECCV 2022 · similarity 0.91)
- Hierarchical Point-based Active Learning for Semi-supervised Point Cloud Semantic Segmentation (ICCV 2023 · similarity 0.90)
- Weakly Supervised Segmentation on Outdoor 4D Point Clouds With Temporal Matching and Spatial Graph Propagation (CVPR 2022 · similarity 0.90)
- Multi-Modality Affinity Inference for Weakly Supervised 3D Semantic Segmentation (AAAI 2024 · similarity 0.89)
- Less Is More: Label Recommendation for Weakly Supervised Point Cloud Semantic Segmentation (AAAI 2024 · 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.
