Research map / Point clouds / Shape and surface completion
AEDNet: Adaptive Embedding and Multiview-Aware Disentanglement for Point Cloud Completion
Zhiheng Fu, Longguang Wang, Lian Xu, Zhiyong Wang, Hamid Laga, Yulan Guo, Farid Boussaid, Mohammed Bennamoun
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 |
|---|---|---|---|
| use AED feature extraction instead of the SnowflakeNet feature extractor (ablation) | AEDNet · MVP (2048 points) | L2 CD x10^4: 5.38 -> 5.12 | Improved |
| use embedding-based coarse completion instead of SnowflakeNet's (ablation) | AEDNet · MVP (2048 points) | L2 CD x10^4: 5.43 -> 5.12 | Improved |
| use GAED fine completion instead of snowflake point deconvolution (ablation) | AEDNet · MVP (2048 points) | L2 CD x10^4: 5.54 -> 5.12 | Improved |
| use multiview-aware disentanglement instead of deconvolution (ablation) | AEDNet · MVP (2048 points) | L2 CD x10^4: 5.44 -> 5.12 | Improved |
| add dictionary mapping initialization and slot interaction to slot attention (ablation) | AEDNet embedding · MVP (2048 points) | L2 CD x10^4: 5.45 -> 5.12 | Improved |
| add dictionary mapping initialization only (ablation) | slot attention · MVP (2048 points) | F-Score@1%: 0.488 -> 0.502 | Improved |
| add slot self-attention interaction only (ablation) | slot attention · MVP (2048 points) | F-Score@1%: 0.488 -> 0.510 | Improved |
| AEDNet versus strongest prior method at full resolution | AEDNet · MVP (16384 points) | L2 CD x10^4: 3.12 -> 2.24 | Improved |
| AEDNet versus Anchorformer on PCN | AEDNet · PCN | L1 CD x10^3: 6.59 -> 6.52 | Improved |
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.
- Anchorformer — AnchorFormer: Point Cloud Completion From Discriminative Nodes (CVPR 2023) · better on 1 of 1 shared result
- SnowflakeNet — SnowflakeNet: Point Cloud Completion by Snowflake Point Deconvolution With Skip-Transformer (ICCV 2021) · 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
- Point Cloud Completion by Skip-Attention Network With Hierarchical Folding (CVPR 2020 · similarity 0.90)
- SoftPool++: An Encoder–Decoder Network for Point Cloud Completion (IJCV 2022 · similarity 0.89)
- Orthogonal Dictionary Guided Shape Completion Network for Point Cloud (AAAI 2024 · similarity 0.88)
- DeCoTR: Enhancing Depth Completion with 2D and 3D Attentions (CVPR 2024 · similarity 0.88)
- DC-PCN: Point Cloud Completion Network with Dual-Codebook Guided Quantization (AAAI 2025 · similarity 0.87)
- Position-Aware Guided Point Cloud Completion with CLIP Model (AAAI 2025 · similarity 0.87)
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.
