Notice
Recent Posts
Recent Comments
Link
| 일 | 월 | 화 | 수 | 목 | 금 | 토 |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | ||||
| 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| 11 | 12 | 13 | 14 | 15 | 16 | 17 |
| 18 | 19 | 20 | 21 | 22 | 23 | 24 |
| 25 | 26 | 27 | 28 | 29 | 30 | 31 |
Tags
- mmcv
- learning to prompt
- Class Incremental Learning
- state_dict()
- Face Pose Estimation
- Markov transition matrix
- Mask diffusion
- CIL
- L2P
- 베이지안 정리
- VQ-VAE
- Img2pose
- prompt learning
- CVPR2022
- timm
- VQ-diffusion
- requires_grad
- Discrete diffusion
- Continual Learning
- Facial Landmark Localization
- Energy-based model
- Vector Quantized Diffusion Model for Text-to-Image Synthesis
- ENERGY-BASED MODELS FOR CONTINUAL LEARNING
- PnP algorithm
- Mask-and-replace diffusion strategy
- Class Incremental
- Face Alignment
- DualPrompt
- img2pose: Face Alignment and Detection via 6DoF
- learning to prompt for continual learning
Archives
- Today
- Total
Computer Vision , AI
[One-page summary] TuneA Video: One Shot Tuning of Image Diffusion Models for Text to Video Generation by Wu et al. 본문
Paper_review[short]
[One-page summary] TuneA Video: One Shot Tuning of Image Diffusion Models for Text to Video Generation by Wu et al.
Elune001 2024. 1. 16. 00:27● Summary: Text to Video generation model using Text to Image diffusion model
● Approach highlight
-
Spatio-temporal attention for efficiency: attend to selected previous frame( first, previous frame)
-
T2V generation using T2I model fine-tuning: update only attention block in fine-tuning stage
● Main Results
● Discussion
- lack of ability to represent multiple object interactions due to limitations of the underlying diffusion model
'Paper_review[short]' 카테고리의 다른 글
'Paper_review[short]' Related Articles
more
