Research map / Video / Video segmentation and tracking
MINI-Net: Multiple Instance Ranking Network for Video Highlight Detection
Fa-Ting Hong, Xuanteng Huang, Wei-Hong Li, Wei-Shi Zheng
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 bag classification module (ablation) | MINI-Net · TVSum | mAP: 65.58% -> 73.24% | Improved |
| add max-max ranking loss (ablation) | MINI-Net · CoSum | mAP: 77.59% -> 92.78% | Improved |
| add audio feature to vision (ablation) | MINI-Net · YouTube Highlights | mAP: 0.6138 -> 0.6436 | Improved |
| add audio feature to vision (ablation) | MINI-Net · TVSum | mAP: 0.6979 -> 0.7324 | Improved |
| add audio feature to vision (ablation) | MINI-Net · CoSum | mAP: 0.8605 -> 0.9278 | Improved |
| add vision feature to audio-only model (ablation) | MINI-Net · CoSum | mAP: 0.6914 -> 0.9278 | Improved |
| use max-max instead of min-min ranking loss (ablation) | MINI-Net · CoSum | mAP: 0.7863 -> 0.9278 | Improved |
| use max-max instead of min-max ranking loss (ablation) | MINI-Net · TVSum | mAP: 0.6747 -> 0.7324 | Improved |
| use max-max instead of max-min ranking loss (ablation) | MINI-Net · YouTube Highlights | mAP: 0.6186 -> 0.6436 | Improved |
| use MINI-Net instead of Gated-Attention MIL | MINI-Net · CoSum | mAP: 0.7516 -> 0.9278 | Improved |
| use MINI-Net instead of DMIL-RM MIL | MINI-Net · CoSum | mAP: 0.7943 -> 0.9278 | Improved |
| vary video length threshold tau from 40 to 60 (ablation) | MINI-Net · CoSum | mAP: 0.8622 -> 0.9278 | Improved |
Later papers that compared with this method
2 later accepted papers report MINI-Net as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Learning Pixel-Level Distinctions for Video Highlight Detection (CVPR 2022 · better on 1 of 1 shared result)
- Temporal Cue Guided Video Highlight Detection With Low-Rank Audio-Visual Fusion (ICCV 2021 · better on 2 of 2 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
- Less Is More: Learning Highlight Detection From Video Duration (CVPR 2019 · similarity 0.90)
- Adaptive Video Highlight Detection by Learning from User History (ECCV 2020 · similarity 0.90)
- Temporal Cue Guided Video Highlight Detection With Low-Rank Audio-Visual Fusion (ICCV 2021 · similarity 0.89)
- Contrastive Learning for Unsupervised Video Highlight Detection (CVPR 2022 · similarity 0.89)
- Find Objects and Focus on Highlights: Mining Object Semantics for Video Highlight Detection via Graph Neural Networks (AAAI 2020 · similarity 0.89)
- Minimizing Labeled, Maximizing Unlabeled: An Image-Driven Approach for Video Instance Segmentation (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.
