Research map / Video / Video generation
First Frame Is the Place to Go for Video Content Customization
Jingxi Chen, Zongxia Li, Zhichao Liu, Guangyao Shi, Xiyang Wu, Fuxiao Liu, Cornelia Fermüller, Brandon Y. Feng, Yiannis Aloimonos
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 few-shot FFGo LoRA adaptation (50 examples) | Wan2.2-I2V-A14B · User study (50 test sets, 40 users) | overall quality: 2.09 -> 4.28 | Improved |
| add few-shot FFGo LoRA adaptation | Wan2.2-I2V-A14B · User study (50 test sets, 40 users) | object identity: 3.32 -> 4.53 | Improved |
| add few-shot FFGo LoRA adaptation | Wan2.2-I2V-A14B · User study (50 test sets, 40 users) | scene identity: 3.01 -> 4.58 | Improved |
| add few-shot FFGo LoRA adaptation | Wan2.2-I2V-A14B · User study (50 test sets, 40 users) | average rank: 3.27 -> 1.21 | Improved |
| add few-shot FFGo LoRA adaptation | Wan2.2-I2V-A14B · User study (50 test sets, 40 users) | ranked 1st (%): 3.4% -> 81.2% | Improved |
| use 50-example LoRA instead of architecture-modified VACE trained on millions of videos | Wan2.2-I2V-A14B · User study (50 test sets, 40 users) | overall quality: 3.00 -> 4.28 | Improved |
| use FFGo instead of SkyReels-A2 | Wan2.2-I2V-A14B · User study (50 test sets, 40 users) | overall quality: 2.34 -> 4.28 | Improved |
| add FFGo LoRA adaptation | Wan2.2-I2V-A14B · Automatic metrics (50 test sets) | Dynamic Degree: 9.75 -> 14.64 | Improved |
| add FFGo LoRA adaptation | Wan2.2-I2V-A14B · Automatic metrics (50 test sets) | DINO-I: 0.42 -> 0.46 | Improved |
| add FFGo LoRA adaptation | Wan2.2-I2V-A14B · Automatic metrics (50 test sets) | Imaging Quality: 0.61 -> 0.73 | Improved |
| use FFGo instead of VACE on image-based concept fidelity | Wan2.2-I2V-A14B · Automatic metrics (50 test sets) | CLIP-I: 0.68 -> 0.67 | No clear effect |
| use FFGo instead of VACE on aesthetics | Wan2.2-I2V-A14B · Automatic metrics (50 test sets) | Aesthetic Quality: 0.92 -> 0.85 | Hurt |
Closest papers by meaning
- VideoMage: Multi-Subject and Motion Customization of Text-to-Video Diffusion Models (CVPR 2025 · similarity 0.87)
- Generative Video Propagation (CVPR 2025 · similarity 0.86)
- CustomCrafter: Customized Video Generation with Preserving Motion and Concept Composition Abilities (AAAI 2025 · similarity 0.86)
- VMC: Video Motion Customization using Temporal Attention Adaption for Text-to-Video Diffusion Models (CVPR 2024 · similarity 0.85)
- VBench: Comprehensive Benchmark Suite for Video Generative Models (CVPR 2024 · similarity 0.85)
- VIRES: Video Instance Repainting via Sketch and Text Guided Generation (CVPR 2025 · 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.
