Research map / Video / Video generation
RealGeneral: Unifying Visual Generation via Temporal In-Context Learning with Video Models
Yijing Lin, Mengqi Huang, Shuhan Zhuang, Zhendong Mao
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 Unified Conditional Embedding (ablation) | RealGeneral · DreamBench customization | CLIP-I: 0.833 -> 0.849 | Improved |
| add Separated Condition AdaLN (ablation) | RealGeneral · DreamBench customization | CLIP-I: 0.830 -> 0.849 | Improved |
| add Frame-Condition Decoupling (ablation) | RealGeneral · DreamBench customization | CLIP-I: 0.827 -> 0.849 | Improved |
| add Unified Conditional Embedding (ablation) | RealGeneral · DreamBench customization | CLIP-T: 0.320 -> 0.314 | Hurt |
| add Frame-Condition Decoupling (ablation) | RealGeneral · DreamBench customization | DINO: 0.665 -> 0.668 | Improved |
| use Mask A (block text-condition image interaction) instead of no mask (ablation) | RealGeneral · DreamBench customization | CLIP-I: 0.827 -> 0.849 | Improved |
| use Mask A instead of Mask C (ablation) | RealGeneral · DreamBench customization | CLIP-I: 0.779 -> 0.849 | Improved |
| use Mask A instead of Mask B (ablation) | RealGeneral · DreamBench customization | CLIP-T: 0.321 -> 0.314 | Hurt |
| use CogVideoX video base instead of SDXL image base (vs EMU2) | RealGeneral · DreamBench customization | DINO: 0.583 -> 0.668 | Improved |
| use RealGeneral instead of OminiControl | RealGeneral · COCO2017 canny-to-image | FID: 20.63 -> 17.50 | Improved |
| use RealGeneral instead of OmniGen | RealGeneral · COCO2017 depth-to-image | FID: 21.62 -> 23.40 | Hurt |
Closest papers by meaning
- DuoGen: Towards Autonomous Interleaved Multimodal Generation (CVPR 2026 · similarity 0.89)
- Instruct-Imagen: Image Generation with Multi-modal Instruction (CVPR 2024 · similarity 0.88)
- Learning Visual Generative Priors without Text (CVPR 2025 · similarity 0.88)
- Video-Bench: Human-Aligned Video Generation Benchmark (CVPR 2025 · similarity 0.88)
- Janus-Pro-R1: Advancing Collaborative Visual Comprehension and Generation via Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Generative Inbetweening through Frame-wise Conditions-Driven Video Generation (CVPR 2025 · similarity 0.88)
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.
