Research map / Video / Video segmentation and tracking
A Spherical Convolution Approach for Learning Long Term Viewport Prediction in 360 Immersive Video
Chenglei Wu, Ruixiao Zhang, Zhi Wang, Lifeng Sun
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 near-identity kernel instead of equatorial kernel (ablation) | S2CNN feature extractor · PanoUCF101 | top-1 accuracy (%): 57.9 -> 65.7 | Improved |
| add one more SO3Conv layer (6 layers) (ablation) | S2CNN feature extractor · PanoUCF101 | top-1 accuracy (%): 65.7 -> 76.5 | Improved |
| increase frame sampling rate F from 4 to 16 (ablation) | S2CNN+ConvGRU · PanoUCF101 | top-1 accuracy (%): 70.3 -> 73.2 | Improved |
| remove preference and attention (History only) vs AttPref (ablation) | AttPref viewport predictor · 360 viewport dataset (Wu et al. 2017) | viewport accuracy (IOU): 16.2% reduction at wk=8 without them | Improved |
| replace 360Sal saliency features with AttPref | viewport predictor · 360 viewport dataset (Wu et al. 2017) | viewport accuracy (IOU): 32.3% improvement for wk=6 | Improved |
Closest papers by meaning
- Viewport-based CNN: A Multi-task Approach for Assessing 360° Video Quality (TPAMI 2020 · similarity 0.87)
- Learning Spherical Convolution for Fast Features from 360° Imagery (NeurIPS 2017 · similarity 0.87)
- Imagine360: Immersive 360 Video Generation from Perspective Anchor (NeurIPS 2025 · similarity 0.83)
- DiVa-360: The Dynamic Visual Dataset for Immersive Neural Fields (CVPR 2024 · similarity 0.83)
- From an Image to a Scene: Learning to Imagine the World from a Million 360° Videos (NeurIPS 2024 · similarity 0.82)
- DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Vision (CVPR 2024 · similarity 0.82)
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.
