Research map / Video / Video segmentation and tracking
Self-View Grounding Given a Narrated 360° Video
Shih-Han Chou, Yi-Chun Chen, Kuo-Hao Zeng, Hou-Ning Hu, Jianlong Fu, Min 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 |
|---|---|---|---|
| Embed the NFoV generator in feature space (ablation) | RL · Narrated 360° videos, indoor | avg. Recall: 6.1 -> 8.3 | Improved |
| Embed the NFoV generator in feature space (ablation) | RL · Narrated 360° videos, indoor | avg. Precision: 12.6 -> 19.2 | Improved |
| Add 360° data augmentation (ablation) | RL-f · Narrated 360° videos, indoor | avg. Recall: 8.3 -> 12.8 | Improved |
| Add 360° data augmentation (ablation) | RL-f · Narrated 360° videos, indoor | avg. Precision: 19.2 -> 27.8 | Improved |
| Add irrelevant loss (ablation) | D†-RL-f · Narrated 360° videos, indoor | avg. Recall: 12.8 -> 13.4 | Improved |
| Add irrelevant loss (ablation) | D†-RL-f · Narrated 360° videos, indoor | avg. Precision: 27.8 -> 30.7 | Improved |
| Use D†-RIL-f instead of centric selection | NFoV grounding · Narrated 360° videos, total dataset | avg. Recall: 9.7 -> 18.1 | Improved |
| Use D†-RIL-f instead of centric selection | NFoV grounding · Narrated 360° videos, total dataset | avg. Precision: 28.2 -> 34.6 | Improved |
Closest papers by meaning
- SimVG: A Simple Framework for Visual Grounding with Decoupled Multi-modal Fusion (NeurIPS 2024 · similarity 0.84)
- Exploiting Auxiliary Caption for Video Grounding (AAAI 2024 · similarity 0.84)
- Comprehensive Visual Grounding for Video Description (AAAI 2024 · similarity 0.84)
- Referencing Where to Focus: Improving Visual Grounding with Referential Query (NeurIPS 2024 · similarity 0.84)
- SnAG: Scalable and Accurate Video Grounding (CVPR 2024 · similarity 0.83)
- Viewpoint-Aware Visual Grounding in 3D Scenes (CVPR 2024 · similarity 0.83)
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.
