Research map / 3D vision / Depth and stereo
RigNet++: Semantic Assisted Repetitive Image Guided Network for Depth Completion
Zhiqiang Yan, Xiang Li, Le Hui, Zhenyu Zhang, Jun Li, Jian Wu Yang
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 SAM semantic prior to repetitive guidance (ablation) | RigNet++ RG · KITTI val | RMSE (mm): 757.4 -> 752.8 | Improved |
| add RASPN region-aware propagation (ablation) | RigNet++ (j) · KITTI val | RMSE (mm): 743.5 -> 734.7 | Improved |
| add semantic prior to the RASPN variant (full model) (ablation) | RigNet++ (k) · KITTI val | RMSE (mm): 734.7 -> 728.2 | Improved |
| replace guided convolution G1 with efficient guidance EG1 (ablation) | GuideNet baseline with RHN-R18 · KITTI val | RMSE (mm) / memory (GB): 769.0 -> 768.6; memory +1.35 -> -10.60 | Improved |
| increase EG repetitions from 1 to 3 (ablation) | RigNet++ RG · KITTI val | RMSE (mm): 768.6 -> 762.3 -> 757.4 | Improved |
| add adaptive fusion over repetition outputs (ablation) | RigNet++ RG · KITTI val | RMSE (mm): 757.4 -> 752.1 | Improved |
| use our AF instead of simple addition (ablation) | RigNet++ fusion · KITTI val | RMSE (mm): 755.8 -> 752.1 | Improved |
| replace ResNet-18 with EfficientNet-B0 backbone (ablation) | RigNet++ RHN · KITTI val | RMSE (mm): 752.1 -> 748.2 | Improved |
| add dense connection across RHNs (ablation) | RigNet++ · KITTI val | RMSE (mm): 748.2 -> 743.5 | Improved |
| deepen single backbone from ResNet-18 to ResNet-34/50 (ablation) | Deeper single backbone · KITTI val | RMSE (mm): 779 -> 778 -> 777 | No clear effect |
| stack hourglass units in tandem (18-2) (ablation) | Deeper-18-1 · KITTI val | RMSE (mm): 779 -> 802 | Hurt |
| use parallel repetitive RHN-18-3 instead of single ResNet-18 (ablation) | Deeper-18-1 · KITTI val | RMSE (mm): 779 -> 769 | 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.
- RigNet — RigNet: Repetitive Image Guided Network for Depth Completion (ECCV 2022) · better on 5 of 5 shared results
- BEV@DC — BEV@DC: Bird's-Eye View Assisted Training for Depth Completion (CVPR 2023) · better on 4 of 4 shared results
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
- RigNet: Repetitive Image Guided Network for Depth Completion (ECCV 2022 · similarity 0.97)
- CSPN++: Learning Context and Resource Aware Convolutional Spatial Propagation Networks for Depth Completion (AAAI 2020 · similarity 0.90)
- GraphCSPN: Geometry-Aware Depth Completion via Dynamic GCNs (ECCV 2022 · similarity 0.89)
- Dynamic Spatial Propagation Network for Depth Completion (AAAI 2022 · similarity 0.89)
- Completion as Enhancement: A Degradation-Aware Selective Image Guided Network for Depth Completion (CVPR 2025 · similarity 0.88)
- Bilateral Propagation Network for Depth Completion (CVPR 2024 · 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.
