Research map / Label-efficient and robust learning / Self-supervised and contrastive
Shot Contrastive Self-Supervised Learning for Scene Boundary Detection
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Replace SpecAugment positives with shot-similarity positives (ablation) | ShotCoL · AdCuepoints, 10 shot setting | AP: 50.53 -> 53.27 | Helped |
| Add provided pre-computed audio shot features to visual keyframe features (ablation) | ShotCoL scene boundary detection · MovieNet | AP: 52.34 -> 52.47 | Helped |
| Use raw audio to learn audio features rather than relying on pre-computed audio features (ablation) | ShotCoL scene boundary detection · MovieNet | AP | Mixed |
| Decay the learning rate after training performance saturates (ablation) | ShotCoL visual contrastive learning · MovieNet | top-1 accuracy | Helped |
| Initialize the visual encoder randomly instead of using ImageNet-pretrained weights (ablation) | ShotCoL visual contrastive learning · MovieNet | contrastive loss | No clear effect |
Closest papers by meaning
- Video Scene Segmentation with Genre and Duration Signals (ICLR 2026 · similarity 0.88)
- Towards Global Video Scene Segmentation with Context-Aware Transformer (AAAI 2023 · similarity 0.88)
- Neighbor Relations Matter in Video Scene Detection (CVPR 2024 · similarity 0.85)
- Modality-Aware Shot Relating and Comparing for Video Scene Detection (AAAI 2025 · similarity 0.84)
- ShotBench: Expert-Level Cinematic Understanding in Vision-Language Models (NeurIPS 2025 · similarity 0.83)
- Beyond Pixel and Object: Part Feature as Reference for Few-Shot Video Object Segmentation (AAAI 2025 · 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.