Research map / Point clouds / Shape and surface completion
FBNet: Feedback Network for Point Cloud Completion
Xuejun Yan, Hongyu Yan, Jingjing Wang, Hang Du, Zhihong Wu, Di Xie, Shiliang Pu, Li Lu
Read the paperCheck 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 feedback connections (T=3) (ablation) | FBNet · MVP (2048 points) | Chamfer distance: 5.33 -> 5.06 | Improved |
| add feedback connections (T=2) (ablation) | FBNet · MVP (2048 points) | Chamfer distance: 5.36 -> 5.19 | Improved |
| unroll over 3 time steps instead of 1 (with feedback) (ablation) | FBNet · MVP (2048 points) | Chamfer distance: 5.41 -> 5.06 | Improved |
| add time steps without feedback (T=3 vs T=1) (ablation) | FBNet · MVP (2048 points) | Chamfer distance: 5.41 -> 5.33 | No clear effect |
| add partial input to other blocks' initialization (Pff+Ppart instead of Pff) (ablation) | FBNet · MVP (2048 points) | Chamfer distance: 5.31 -> 5.06 | Improved |
| use feedback points as first-block initialization instead of coarse Pc (ablation) | FBNet · MVP (2048 points) | Chamfer distance: 5.28 -> 5.06 | Improved |
| merge feedback points with feedforward points in other blocks instead of partial input (ablation) | FBNet · MVP (2048 points) | Chamfer distance: 5.06 -> 19.93 | Hurt |
| use AdaptGP instead of graph pooling (ablation) | HGNet · MVP (2048 points) | Chamfer distance: 5.48 -> 5.06 | Improved |
| use AdaptGP instead of point pooling (ablation) | HGNet · MVP (2048 points) | Chamfer distance: 5.23 -> 5.06 | Improved |
| increase time steps from 1 to 2 (ablation) | FBNet · MVP (16384 points) | Chamfer distance: 2.59 -> 2.29 | Improved |
| later time step output (t=2 vs t=0) (ablation) | FBNet · MVP (2048 points) | Chamfer distance: 5.79 -> 5.06 | Improved |
| replace SnowflakeNet with FBNet | SnowflakeNet · PCN dataset (16384 points) | L1 Chamfer distance: 7.21 -> 6.94 | Improved |
Later papers that compared with this method
1 later accepted paper reports FBNet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- SymmCompletion: High-Fidelity and High-Consistency Point Cloud Completion with Symmetry Guidance (AAAI 2025 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SnowflakeNet — SnowflakeNet: Point Cloud Completion by Snowflake Point Deconvolution With Skip-Transformer (ICCV 2021) · better on 7 of 7 shared results
- GRNet — GRNet: Gridding Residual Network for Dense Point Cloud Completion (ECCV 2020) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Cascaded Refinement Network for Point Cloud Completion with Self-supervision (TPAMI 2021 · similarity 0.91)
- PMP-Net++: Point Cloud Completion by Transformer-Enhanced Multi-Step Point Moving Paths (TPAMI 2022 · similarity 0.90)
- PointAttN: You Only Need Attention for Point Cloud Completion (AAAI 2024 · similarity 0.89)
- CDPNet: Cross-Modal Dual Phases Network for Point Cloud Completion (AAAI 2024 · similarity 0.89)
- DC-PCN: Point Cloud Completion Network with Dual-Codebook Guided Quantization (AAAI 2025 · similarity 0.89)
- Orthogonal Dictionary Guided Shape Completion Network for Point Cloud (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.
