Selected visual results from a private research prototype exploring temporal consistency and motion coherence for video generation and editing.
Real or real-looking visual demos. · Private implementation. · Public teaser only.
| Demo | Preview | Videos |
|---|---|---|
| Demo 1 | ![]() |
before · after |
| Demo 2 | ![]() |
before · after |
| Demo 3 | ![]() |
before · after |
| Demo 4 | ![]() |
before · after |
| Demo 5 | ![]() |
before · after |
This repository is a public teaser for a private research prototype. It contains selected visual outputs only. The technical implementation is not publicly released.
- Short before/after MP4 demos.
- GIF previews for quick viewing on GitHub.
- A small visual gallery.
- A high-level non-technical summary (RESULTS_SUMMARY_NON_ENABLING.md).
- No source code.
- No formulas.
- No technical diagrams.
- No training details.
- No model checkpoints.
- No private reports.
- No evaluation tables.
- No implementation details.
These demos are early research previews. They are not a product, not a benchmark result, and not a state-of-the-art result. The goal of this public page is simply to show selected visual outputs from an ongoing private research direction.
No advantage of any kind is asserted.
Part of my research portfolio → chen-quantum.github.io/projects
Chen Gadi chengadi@mail.tau.ac.il





