Research map / Video / Video-language understanding
Syntax Customized Video Captioning by Imitating Exemplar Sentences
Yitian Yuan, Lin Ma, Wenwu Zhu
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 |
|---|---|---|---|
| replace exemplar syntax conditioning with the proposed model instead of Seq2Seq | Ours · MSRVTT | TED: 15.91 -> 5.44 | Improved |
| replace exemplar syntax conditioning with the proposed model instead of GFN | Ours · ActivityNet Captions | TED: 20.86 -> 5.52 | Improved |
| use proposed model instead of Template | Ours · MSRVTT | perplexity: 7.75 -> 5.64 | Improved |
| use proposed model instead of Template | Ours · MSRVTT | COS: 0.6595 -> 0.6892 | Improved |
| use proposed model instead of GFN | Ours · MSRVTT | LSA diversity: 0.2194 -> 0.6150 | Improved |
| use proposed model instead of GFN or Seq2Seq | Ours · MSRVTT | CIDEr: 43.90 -> 11.45 | Hurt |
| add exemplar sentence corpus autoencoding loss L_ss (ablation) | Ours · MSRVTT | TED/COS | Improved |
| add character-level LSTM to syntax encoder (ablation) | Ours · MSRVTT | TED/COS | Improved |
| add POS word replacement mechanism (ablation) | Ours · MSRVTT | TED/COS | Improved |
| add Conditional Layer Normalization to decoding LSTM (ablation) | Ours · MSRVTT | COS | Improved |
| put syntax first and semantics second in decoder layers (ablation) | Ours · MSRVTT | TED/COS | Improved |
| raise word replacement probability from 0.1 to 0.7 (ablation) | Ours · MSRVTT | COS: 0.6535 -> 0.6892 | Improved |
Closest papers by meaning
- Joint Syntax Representation Learning and Visual Cue Translation for Video Captioning (ICCV 2019 · similarity 0.89)
- Distilling Vision-Language Models on Millions of Videos (CVPR 2024 · similarity 0.87)
- VideoCon: Robust Video-Language Alignment via Contrast Captions (CVPR 2024 · similarity 0.86)
- Set Prediction Guided by Semantic Concepts for Diverse Video Captioning (AAAI 2024 · similarity 0.86)
- Progress-Aware Video Frame Captioning (CVPR 2025 · similarity 0.86)
- Video ReCap: Recursive Captioning of Hour-Long Videos (CVPR 2024 · 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.
