Research map / 3D vision / 3D generation
Generalizing Visual Geometry Priors to Sparse Gaussian Occupancy Prediction
Changqing Zhou, Yueru Luo, Changhao Chen
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 |
|---|---|---|---|
| increase ray samples K from 1 to 16 (ablation) | GPOcc (VGGT) · Occ-ScanNet (monocular) | mIoU: 47.88 -> 56.19 | Improved |
| increase ray samples K from 16 to 32 (ablation) | GPOcc (VGGT) · Occ-ScanNet (monocular) | mIoU: 56.19 -> 56.72 | No clear effect |
| raise opacity pruning threshold from 0.01 to 0.10 (ablation) | GPOcc (VGGT) · Occ-ScanNet (monocular) | mIoU: 56.19 -> 52.65 | Hurt |
| raise opacity pruning threshold from 0.01 to 0.02 (fewer Gaussians) (ablation) | GPOcc (VGGT) · Occ-ScanNet (monocular) | mIoU: 56.19 -> 55.21 | Hurt |
| replace DepthAnything prior with VGGT prior | GPOcc · Occ-ScanNet (monocular) | mIoU: 51.88 -> 56.19 | Improved |
| replace DepthAnything prior with VGGT prior | GPOcc · EmbodiedOcc-ScanNet (streaming) | mIoU: 51.22 -> 55.39 | Improved |
| use GPOcc ray-sampled sparse Gaussians with DepthAnything instead of EmbodiedOcc | GPOcc (DPT) · Occ-ScanNet (monocular) | mIoU: 45.15 -> 51.88 | Improved |
| use GPOcc-DPT instead of EmbodiedOcc (speed) | GPOcc (DPT) · Occ-ScanNet (monocular) | FPS: 10.66 -> 28.22 | Improved |
| use GPOcc-DPT with incremental update instead of EmbodiedOcc++ | GPOcc (DPT) · EmbodiedOcc-ScanNet (streaming) | mIoU: 43.60 -> 51.22 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- EmbodiedOcc — EmbodiedOcc: Embodied 3D Occupancy Prediction for Vision-based Online Scene Understanding (ICCV 2025) · better on 9 of 9 shared results
- GaussianFormer — GaussianFormer: Scene as Gaussians for Vision-Based 3D Semantic Occupancy Prediction (ECCV 2024) · better on 2 of 2 shared results
- SurroundOcc — SurroundOcc: Multi-camera 3D Occupancy Prediction for Autonomous Driving (ICCV 2023) · better on 2 of 2 shared results
- MonoScene — MonoScene: Monocular 3D Semantic Scene Completion (CVPR 2022) · 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
- ViPOcc: Leveraging Visual Priors from Vision Foundation Models for Single-View 3D Occupancy Prediction (AAAI 2025 · similarity 0.91)
- AdaOcc: Adaptive 3D Occupancy Prediction for Embodied Tasks (NeurIPS 2026 · similarity 0.90)
- ODG: Occupancy Prediction Using Dual Gaussians (NeurIPS 2025 · similarity 0.90)
- GaussianFormer-2: Probabilistic Gaussian Superposition for Efficient 3D Occupancy Prediction (CVPR 2025 · similarity 0.89)
- 3D Occupancy Prediction with Low-Resolution Queries via Prototype-aware View Transformation (CVPR 2025 · similarity 0.88)
- RadOcc: Learning Cross-Modality Occupancy Knowledge through Rendering Assisted Distillation (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.
