Research map / 3D vision / Depth and stereo
A Survey of Representation Learning, Optimization Strategies, and Applications for Omnidirectional Vision
Hao Ai, Zidong Cao, Lin Wang
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 PanoSwin-Tiny instead of Spherical GCN | spherical image classifiers · SPH-CIFAR10 | accuracy (%): 60.72 -> 75.01 | Improved |
| use PanoSwin-Tiny instead of SphereNet | spherical image classifiers · SPH-MNIST | accuracy (%): 94.41 -> 99.85 | Improved |
| use a no-reference learned VQA model (Assessor360) instead of WS-PSNR | ODI quality assessment · OIQA | SRCC: 0.3829 -> 0.9802 | Improved |
| use CIQNet instead of BOVQA | ODV quality assessment · VQA-ODV | SRCC: 0.6403 -> 0.9342 | Improved |
| use NR learned methods VGCN instead of FR OV-PSNR on videos | ODV quality assessment · VQA-ODV | SRCC: 0.7503 -> 0.4092 | Hurt |
| add distortion-aware adaptation (Trans4PASS) over Dense-PASS | UDA panoramic segmentation · DensePASS | mIoU (%): 43.02 -> 55.22 | Improved |
| add SAM-guided pseudo labels (GoodSAM) over DATR | UDA panoramic segmentation · DensePASS | mIoU (%): 56.81 -> 60.56 | Improved |
| add projection-aware adaptation (DPPASS) over Trans4PASS | UDA panoramic segmentation · DensePASS | mIoU (%): 55.22 -> 56.28 | Improved |
| use SGAT4PASS instead of Trans4PASS-Small | supervised panoramic segmentation · Stanford2D3D | Avg mIoU (%): 52.1 -> 55.3 | Improved |
| use transformer PanoFormer instead of tangent ResNet-101 | supervised panoramic segmentation · Stanford2D3D | Avg mIoU (%): 45.6 -> 48.9 | Improved |
Closest papers by meaning
- OmniLocalRF: Omnidirectional Local Radiance Fields from Dynamic Videos (CVPR 2024 · similarity 0.88)
- OmniSDF: Scene Reconstruction using Omnidirectional Signed Distance Functions and Adaptive Binoctrees (CVPR 2024 · similarity 0.87)
- ODGS: 3D Scene Reconstruction from Omnidirectional Images with 3D Gaussian Splattings (NeurIPS 2024 · similarity 0.87)
- ODI-Bench: Can MLLMs Understand Immersive Omnidirectional Environments? (ICLR 2026 · similarity 0.87)
- OmniSplat: Taming Feed-Forward 3D Gaussian Splatting for Omnidirectional Images with Editable Capabilities (CVPR 2025 · similarity 0.87)
- Omni-Scene: Omni-Gaussian Representation for Ego-Centric Sparse-View Scene Reconstruction (CVPR 2025 · similarity 0.86)
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.
