Research map / Vision-language and multimodal / Human and embodied modeling
``Factual'' or ``Emotional'': Stylized Image Captioning with Adaptive Learning and Attention
Tianlang Chen, Zhongping Zhang, Quanzeng You, Chen Fang, Zhaowen Wang, Hailin Jin, Jiebo Luo
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 adaptive learning with reference factual model (ablation) | SF-LSTM · FlickrStyle10K humorous, factual refs | CIDEr: 28.2 -> 34.2 | Improved |
| add adaptive learning with reference factual model (ablation) | SF-LSTM · FlickrStyle10K humorous, factual refs | BLEU-4: 13.1 -> 15.4 | Improved |
| add adaptive learning with reference factual model (ablation) | SF-LSTM · FlickrStyle10K romantic, factual refs | CIDEr: 23.5 -> 26.7 | Improved |
| add adaptive learning with reference factual model (ablation) | SF-LSTM · FlickrStyle10K romantic, factual refs | BLEU-4: 12.8 -> 13.5 | Improved |
| add adaptive learning with reference factual model (ablation) | SF-LSTM · FlickrStyle10K humorous, stylized refs | CIDEr: 39.8 -> 39.5 | No clear effect |
| add adaptive learning with reference factual model (ablation) | SF-LSTM · FlickrStyle10K humorous, stylized refs | BLEU-4: 4.9 -> 5.1 | Improved |
| add adaptive learning with reference factual model (ablation) | SF-LSTM · FlickrStyle10K romantic, stylized refs | CIDEr: 37.4 -> 37.5 | No clear effect |
| add adaptive learning with reference factual model (ablation) | SF-LSTM · FlickrStyle10K romantic, stylized refs | BLEU-4: 4.9 -> 4.8 | No clear effect |
| use style-factual LSTM instead of fine-tuned NIC | SF-LSTM · FlickrStyle10K humorous, stylized refs | CIDEr: 35.4 -> 39.8 | Improved |
| use style-factual LSTM instead of fine-tuned NIC | SF-LSTM · FlickrStyle10K romantic, factual refs | CIDEr: 24.0 -> 23.5 | No clear effect |
| use SF-LSTM + Adap instead of SentiCap | SF-LSTM + Adap · Image sentiment captioning, POS test set | CIDEr: 54.4 -> 60.0 | Improved |
| use SF-LSTM + Adap instead of SentiCap | SF-LSTM + Adap · Image sentiment captioning, NEG test set | CIDEr: 61.8 -> 59.7 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- SentiCap — SentiCap: Generating Image Descriptions with Sentiments (AAAI 2016) · better on 1 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
- StyleNet: Generating Attractive Visual Captions With Styles (CVPR 2017 · similarity 0.90)
- MemCap: Memorizing Style Knowledge for Image Captioning (AAAI 2020 · similarity 0.89)
- SentiCap: Generating Image Descriptions with Sentiments (AAAI 2016 · similarity 0.89)
- MSCap: Multi-Style Image Captioning With Unpaired Stylized Text (CVPR 2019 · similarity 0.88)
- Noise-Aware Image Captioning with Progressively Exploring Mismatched Words (AAAI 2024 · similarity 0.86)
- Image Captioning with Multi-Context Synthetic Data (AAAI 2024 · similarity 0.85)
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.
