First Order Motion Model
Description
A famous deep learning model for image animation (e.g., making a single portrait image speak or move) by modeling motion as a first order approximation. Often used in one-shot talking head video generation.
Resource details
- Category path
- Media Tools
- Classification
- Not yet classified
First Order Motion Model is cataloged in Media Tools.
URL
Related resources
- Encoder-Quantization-Motion (EQM) Video Quality MetricAcademic paper on an encoder-side video quality metric for low-latency use, addressing gaps in VMAF/P.1204.3 via contrast/motion masking an…
- VQA² (Visual Question Answering for Video Quality Assessment)ACM MM 2025 paper and code for VQA²-Assistant, a unified video/image quality scoring and interpretation model.