nihaomiao/CVPR23_LFDM
A latent flow diffusion model that generates videos from input images using optical-flow-based motion estimation.
Not currently ranked — collecting fresh signals.
star history
This repository provides the PyTorch implementation of a CVPR 2023 paper on conditional image-to-video generation. The model uses latent flow diffusion to synthesize realistic videos from single images, incorporating optical flow estimation to model motion patterns. The implementation includes training code and pretrained models for multiple datasets (MUG, MHAD, NATOPS) and supports cross-domain applications such as animating faces from the FaceForensics dataset.
Frequently asked
- What is nihaomiao/CVPR23_LFDM?
- A latent flow diffusion model that generates videos from input images using optical-flow-based motion estimation.
- Is CVPR23_LFDM open source?
- Yes — nihaomiao/CVPR23_LFDM is open source, released under the BSD-2-Clause license.
- What language is CVPR23_LFDM written in?
- nihaomiao/CVPR23_LFDM is primarily written in Python.
- How popular is CVPR23_LFDM?
- nihaomiao/CVPR23_LFDM has 471 stars on GitHub.
- Where can I find CVPR23_LFDM?
- nihaomiao/CVPR23_LFDM is on GitHub at https://github.com/nihaomiao/CVPR23_LFDM.