Research map / Video / Video-language understanding
Automatic Understanding of Image and Video Advertisements
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 |
|---|---|---|---|
| Group 221 symbols into 53 clusters based on synonymy and co-occurrence (ablation) | Image ad symbol prediction · Image ads | F-score: 15.79% -> 26.84% | Helped |
| Add predicted symbol probabilities as features (ablation) | Single-word image ad QA network · Image ad QA, single-word answers | accuracy: 11.48% -> 11.96% | Helped |
| Replace baseline image features with features from symbol-, topic-, and sentiment-fine-tuned networks (ablation) | 30-prototype image ad QA network · Image ad QA, 30 prototype answers | accuracy: 48.45% -> 50% | Helped |
| Train sentiment recognition on general-image Visual Sentiment Ontology data instead of ad data (ablation) | Image ad sentiment recognition · Image ads, sentiment | accuracy: 27.92% -> 6.64% | Hurt |
| Use C3D fc7 features rather than fc6 features (ablation) | Video ad topic and sentiment classification · Video ads | accuracy | Helped |
| Use learned video features rather than predicting the majority class | Video humor classifier · Video ads, humor | accuracy: 58% -> 78.6% | Helped |
| Use learned video features rather than predicting the majority class | Video excitement classifier · Video ads, excitement | accuracy: 60.8% -> 78.2% | Helped |
Closest papers by meaning
- A3: Towards Advertising Aesthetic Assessment (CVPR 2026 · similarity 0.84)
- AI-generated Image Quality Assessment in Visual Communication (AAAI 2025 · similarity 0.84)
- GeoReward: Mitigating Contextual Variable Overestimation in Vision-Language Models for Cross-Market Preference Prediction (ICML 2026 · similarity 0.82)
- Infer Induced Sentiment of Comment Response to Video: A New Task, Dataset and Baseline (NeurIPS 2024 · similarity 0.81)
- A Multimodal Benchmark for Framing of Oil & Gas Advertising and Potential Greenwashing Detection (NeurIPS 2025 · similarity 0.81)
- Uncertain Multimodal Intention and Emotion Understanding in the Wild (CVPR 2025 · similarity 0.81)
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.